🙌 OpenHands: AI-Driven Development
docs(verify-openhands): backfill all existing open bugs (#18334)
Engel Nysta2cba97cc790
diff --git a/.agents/skills/verify-openhands/SKILL.md b/.agents/skills/verify-openhands/SKILL.mdindex fd9fca07fa..f1942a8e36 100644--- a/.agents/skills/verify-openhands/SKILL.md+++ b/.agents/skills/verify-openhands/SKILL.md@@ -121,7 +121,18 @@ control-openhands stop # stops only this run; evidence s`browser reset` (fresh browser profile: first run again), `service stopautomation|agent-server` (backend-down UI), `restart` (same state after abackend restart: persistence and reconnection) and `restart --rotate-key`- (stale session key). `restart` keeps the browser on its page. While a service+ (stale session key). `restart` keeps the browser on its page.+ `browser network-state offline|online|status` disconnects or reconnects the+ browser context without changing backend state or intercepting requests;+ restore `online` before doctor or teardown.+ `browser events --kinds websocket-created,websocket-open,websocket-close,websocket-error --last 100`+ observes real subscriptions: redacted socket/page URLs preserve numeric+ `after_seq`, with event timestamps and frame counts. No frame bodies or+ headers are recorded. `websocket-created` is an attempt; `websocket-open`+ marks the first observed frame, which proves communication rather than an+ exact handshake time. Counts are snapshots at these lifecycle entries.+ Navigate or reload after the daemon starts to capture fresh subscriptions;+ healthy lifecycle entries are omitted from `browser errors`. While a serviceis stopped, a reload replaces the page with the backend-unavailable screen, sodrive backend-down states on the page that was already loaded.`service pause automation` and `service resume automation` queue and release@@ -231,9 +242,13 @@ prerequisite; never substitute a mock and call it a pass.<date>, #<issue>)` note on the bullet that reproduces it, next to theexpected result, or a `Known issue #<issue>` Gotchas entry when no recipe candrive it. Make the note executable with a `known-bug` block where a recipe- can drive it, so `control-openhands bug run <ID>` reports it as an expected- failure (`xfail`) and flags it (`xpass`) the day it stops reproducing.- Passes re-drive these notes once the issue closes.+ can drive it, so `control-openhands bug run <ID>` reports an expected failure+ (`xfail`) and flags it (`xpass`) when the symptom stops reproducing. The+ [existing-bug backfill](references/backfills/2026-10-12-open-bugs.md) records its+ frozen inventory; the [main refresh](references/backfills/2026-10-16-main-refresh.md)+ reconciles later fixes and intake; [bug-backfill.md](references/bug-backfill.md) defines outcomes+ and label reconciliation. Passes re-drive these notes once the issue closes.+ `ready-for-dev` prioritizes checks without excluding older actionable reports.- **Blocked**: name the missing prerequisite (account, entitlement, OS, binary)and the route you attempted. Unreachable is never a pass.diff --git a/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs-checks.json b/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs-checks.jsonnew file mode 100644index 0000000000..e14ad55dd4--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs-checks.json@@ -0,0 +1,1763 @@+{+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "families_observed": [+ "F02",+ "F03",+ "F04",+ "F05",+ "F06",+ "F07",+ "F08",+ "F10",+ "F11",+ "F13",+ "F14",+ "F15",+ "F16",+ "F17",+ "F18",+ "F19",+ "F20",+ "F21",+ "F22",+ "F23",+ "F25",+ "F26",+ "F27"+ ],+ "counts": {+ "pass": 23,+ "fail": 75,+ "blocked": 7+ },+ "checks": [+ {+ "ts": "2026-10-12T06:13:37.698Z",+ "feature": "F07.open-by-url",+ "entry": "Conversation route > idle untitled conversation",+ "result": "pass",+ "expected": "Local terminal conversation uses ordinary 30-second polling",+ "actual": "Eight browser GETs over 210 seconds had 30.008–30.031-second intervals; real title stayed null and status idle; no 3-second polls observed",+ "artifacts": [+ "evidence/F07.open-by-url/16877-local-polling.json",+ "evidence/F07.open-by-url/16877-local-idle.png"+ ],+ "note": "#16877 Local diagnostic only; Cloud untitled IDLE/FINISHED branch needs an authenticated Cloud fixture",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-coordinator-checks"+ },+ {+ "ts": "2026-10-12T09:11:49.506Z",+ "feature": "F03.composer-blocked-without-llm",+ "entry": "Home > named AgentProfile with different active LLM",+ "result": "fail",+ "expected": "Readiness follows the configuration used by implicit launch",+ "actual": "Keyless pinned/key-present active shows not-set-up warning; inverse hides warning and enables Send, but launches keyless active model with launched_agent_profile null and default concurrency1",+ "artifacts": [+ "evidence/F03.composer-blocked-without-llm/16951-empty-pinned-ready-active.png",+ "evidence/F03.composer-blocked-without-llm/16951-ready-pinned-empty-active.png",+ "evidence/F03.composer-blocked-without-llm/16951-launch-readback.json"+ ],+ "note": "#16951 credential-presence/launch-policy check only; disposable refused-loopback configuration, no successful inference",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-coordinator-checks"+ },+ {+ "ts": "2026-10-12T09:28:37.924Z",+ "feature": "F03.selection-after-reload",+ "entry": "Home > selected Local workspace > SPA backend switch",+ "result": "fail",+ "expected": "Clear incompatible launch target before submission",+ "actual": "QA_Secondary workspace stayed visible after switching to Local; actual Local conversation used QA_Secondary fixture path",+ "artifacts": [+ "evidence/F03.selection-after-reload/16950-target-after-switch.png",+ "evidence/F03.selection-after-reload/16950-submission-readback.json"+ ],+ "note": "#16950 Local variant only; no Cloud organization/repository claim",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-coordinator-checks"+ },+ {+ "ts": "2026-10-12T09:28:38.044Z",+ "feature": "F03.workspace-dropdown-search",+ "entry": "Home > saved workspace picker > SPA Local backend switch",+ "result": "fail",+ "expected": "Secondary picker shows only its own QA_B_ONLY registry",+ "actual": "After warming Local QA_A_ONLY, secondary picker still offered QA_A_ONLY within2.7seconds; actual secondary registry contained only QA_B_ONLY",+ "artifacts": [+ "evidence/F03.workspace-dropdown-search/16844-local-switch.mp4",+ "evidence/F03.workspace-dropdown-search/16844-local-b-picker.png",+ "evidence/F03.workspace-dropdown-search/16844-saved-registry-readback.json"+ ],+ "note": "#16844 saved-workspace cache reproduced; shared-host parent-directory variant and Windows untested",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-coordinator-checks"+ },+ {+ "ts": "2026-10-12T06:17:46.592Z",+ "feature": "F26.session-key-rotated",+ "entry": "OpenHands/OpenHands#18301 — [Bug]: Preserve Automation KV encryption across session-key rotation",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Wrote and read an owned Automation KV marker (HTTP 201/200), then control-openhands restart --rotate-key with no AUTOMATION_KV_SECRET override. Fresh-auth KV GET now returns HTTP 500 instead of the prior marker.",+ "artifacts": [+ "evidence/F26.session-key-rotated/kv-rotation-18301.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:46.711Z",+ "feature": "F26.runtime-services",+ "entry": "OpenHands/OpenHands#18299 — [Bug]: migration 029 errors with `duplicate column name: consecutive_auth_failures` on SQLite",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Stopped only the owned automation service, backed up its DB, reset alembic_version from 032 to 028 while consecutive_auth_failures exists, restarted the real launcher. Migration duplicate-column signature and automation unavailability occur while launcher stays alive. Private original DB restored afterward.",+ "artifacts": [+ "evidence/F26.runtime-services/partial-schema-18299.json",+ "evidence/F26.runtime-services/partial-schema-18299.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:46.817Z",+ "feature": "F20.git-tree-url",+ "entry": "OpenHands/OpenHands#17783 — [Bug]: Canvas App install form throws on malformed percent escapes in tree URLs",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Executed both malformed ref and path tree URLs in the live install modal; each raised URI malformed, no install mutation or toast, modal remains. Reviewed local demo control installs and enables.",+ "artifacts": [+ "evidence/F20.git-tree-url/malformed-ref-17783.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:46.925Z",+ "feature": "F23.script-section",+ "entry": "OpenHands/OpenHands#17704 — [Bug]: A non-ASCII file name in a script bundle shows garbled (café.py as café.py) and loses its Entrypoint badge",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Owned UTF-8 USTAR bundle café.py/main.py, inactive automation python3 café.py: actual Script header renders café.py and no Entrypoint badge; ASCII control untouched.",+ "artifacts": [+ "evidence/F23.script-section/unicode-17704.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.028Z",+ "feature": "F21.phone",+ "entry": "OpenHands/OpenHands#17564 — [Bug]: Mobile automation dashboard actions squeeze the heading to about 52px",+ "result": "pass",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "With 25 inert automation fixtures, the dashboard heading measures 347 px at 390 px and 277 px at 320 px, at 28 px height. Actions stack below it and no horizontal page overflow appears. This exact current dashboard layout does not reproduce the reported 52 px squeeze.",+ "artifacts": [+ "evidence/F21.phone/header-390-17564.png",+ "evidence/F21.phone/header-320-17564.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.130Z",+ "feature": "F26.cli-port-in-use",+ "entry": "OpenHands/OpenHands#17546 — [Bug]: detect ports automatically instead of always using 3001",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. It provides no automatic frontend fallback, and the ingress override does not change that internal port.",+ "artifacts": [+ "evidence/F26.cli-port-in-use/frontend-default-17546-16226.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.231Z",+ "feature": "F25.add-agent-server",+ "entry": "OpenHands/OpenHands#17371 — [Bug]: rotated session key isn't reconciled into openhands-backends → silent permanent 401 loop",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Added QA_Rotation_Legacy for the identical origin through Add Backend using the owned session-key file. After restart --rotate-key and reload, read-only comparison shows default-local matches injected key but the manually registered active entry does not. UI shows Disconnected and Unauthorized; core unreconciled credential persists, although there is visible status feedback in this version.",+ "artifacts": [+ "evidence/F25.recovery-gate/rotated-manual-backend-17371.png",+ "evidence/F25.recovery-gate/rotated-key-booleans-17371.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.337Z",+ "feature": "F21.kebab-menu",+ "entry": "OpenHands/OpenHands#17128 — [Bug]: automation menu is cut off and some options are unreachable",+ "result": "pass",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "The 25th list row is scrolled into view and its action menu is opened at desktop 1440×1000 and phone 390×844. The full 160×236 menu and its 150×36 Delete button fit within both viewports; no option is cut off in this setup.",+ "artifacts": [+ "evidence/F21.phone/last-row-kebab-17128.png",+ "evidence/F21.kebab-menu/last-row-phone-17128.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.442Z",+ "feature": "F23.activity-log",+ "entry": "OpenHands/OpenHands#17121 — [Bug]: Live run phases never render against automation 1.9.0+ (current_phase)",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Actual bounded local script run API current_phase=Starting automation, no structured phase fields; Activity Log reads only Running and contains no phase text.",+ "artifacts": [+ "evidence/F23.activity-log/current-phase-17121.aria.txt",+ "evidence/F23.run-status-polling/missing-phase-17121.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.545Z",+ "feature": "F20.rail-entry",+ "entry": "OpenHands/OpenHands#16995 — [Bug]: Pin as Home is missing from Canvas Extension pages",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Installed and enabled reviewed local demo; hover extension sidebar entry reveals no pin control, while hover Automate reveals sidebar-pin-home-toggle-automations.",+ "artifacts": [+ "evidence/F20.rail-entry/extension-no-home-pin-16995.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.650Z",+ "feature": "F26.runtime-services",+ "entry": "OpenHands/OpenHands#16927 — [Bug]: Launcher stays alive after a required service dies with EADDRINUSE",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Killed only the owned required static frontend after a healthy startup. The genuine launcher remains alive and authenticated /api/settings still returns 200; UI / returns 502 Bad Gateway. This exercises the required-child fatal-exit branch using SIGKILL, not an EADDRINUSE race; exact trigger is recorded. Normal stack restored afterward.",+ "artifacts": [+ "evidence/F26.runtime-services/required-frontend-crash-16927.json",+ "evidence/F26.runtime-services/required-frontend-crash-16927.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.752Z",+ "feature": "F26.runtime-services",+ "entry": "OpenHands/OpenHands#16839 — [Bug]: Proxy timeout listeners accumulate on keep-alive connections",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Twenty sequential genuine ingress authenticated /api/conversations/search requests over one keep-alive connection, allHTTP 200; the owned ingress emitted MaxListenersExceededWarning for11 timeout listeners.",+ "artifacts": [+ "evidence/F26.runtime-services/keepalive-16839.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.856Z",+ "feature": "F25.manage-backends",+ "entry": "OpenHands/OpenHands#16492 — Backend API keys live in origin-readable localStorage — any same-origin script can read all of them",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Read-only same-origin browser inspection records the number of registered backend entries and nonempty plaintext apiKey values; the retained probe now shows both owned entries are readable. No credential value or whole storage JSON was printed.",+ "artifacts": [+ "evidence/F25.manage-backends/plaintext-key-counts-16492.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:47.959Z",+ "feature": "F26.runtime-services",+ "entry": "OpenHands/OpenHands#16383 — Session API key passed as plaintext CLI arg to static-server.mjs (visible via ps/proc)",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Inspect only the owned static-server process for frontend18813: session-key equals a literal argv value. Evidence records booleans only; no credential/argv string exposed.",+ "artifacts": [+ "evidence/F26.runtime-services/argv-16383.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.061Z",+ "feature": "F25.selector-dropdown",+ "entry": "OpenHands/OpenHands#16297 — [Bug]: Default \"Local\" backend name is confusing when on a VM",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Served the unchanged UI on an owned protected non-loopback LAN origin, with auth-required and no injected key. Onboarding default Host Name is Local despite the LAN URL; after authenticating with only the private run key and skipping remaining setup, the actual selector reads Connected Local. This verifies the non-loopback naming branch, not a VM installation.",+ "artifacts": [+ "evidence/F25.selector-dropdown/nonloopback-label-final-16297.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.164Z",+ "feature": "F22.template-card",+ "entry": "OpenHands/OpenHands#16277 — Automate tab crashes on catalog entry with missing `requires` (guard getIntegrationIds)",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Frozen-source isolated copy with only private @openhands/extensions custom-automation manifest.requires removed; build served correctly from build/ against the healthy owned backend. Actual /automations/templates renders Application Error with TypeError: Cannot read properties of undefined (reading integrations), crashing the entire catalog instead of isolating the malformed entry. Shared source/build untouched.",+ "artifacts": [+ "evidence/F22.template-card/missing-requires-16277.png",+ "evidence/F22.template-card/missing-requires-16277.aria.txt"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.267Z",+ "feature": "F26.cli-port-in-use",+ "entry": "OpenHands/OpenHands#16226 — [Bug]: FIXED frontend port results in port conflict",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. The explicit ingress override does not choose another frontend port.",+ "artifacts": [+ "evidence/F26.cli-port-in-use/frontend-default-17546-16226.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.368Z",+ "feature": "F26.runtime-services",+ "entry": "OpenHands/OpenHands#16217 — [Bug]: npm run dev can launch Uvicorn outside the automation environment",+ "result": "pass",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "Installed real uvicorn 0.37.0 alone in an owned Python 3.12 venv, prepended that bin to the private launcher PATH, then restarted the genuine current launcher. Automation 1.19.3 starts and /api/automation/health returns HTTP 200; the reported external-uvicorn ModuleNotFoundError does not occur in this exact resolver/PATH setup. Original PATH restored and doctor run afterward.",+ "artifacts": [+ "evidence/F26.runtime-services/standalone-uvicorn-16217.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.470Z",+ "feature": "F21.list-row",+ "entry": "OpenHands/OpenHands#16155 — [Bug]: \"Run now\" button renders incorrectly when the automation title is long",+ "result": "pass",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "The exact reported long title is arranged in a real inactive list row and scrolled into view. Run now is an icon-only 32×32 px control, with no wrapping Run/now text, in the current list layout.",+ "artifacts": [+ "evidence/F21.list-row/long-title-visible-16155.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.572Z",+ "feature": "F21.add-menu",+ "entry": "OpenHands/OpenHands#15833 — [Bug]: Missing text chat when Create Automation button is clicked in modal",+ "result": "pass",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "Add Automation → Create Automation opens /automations/setup with a New Automation draft editor, prompt, schedule and assistance input. The old modal-to-empty-conversation branch is replaced on this target, so its missing prefilled chat text does not reproduce.",+ "artifacts": [+ "evidence/F22.agent-assisted-setup/create-flow-15833.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.675Z",+ "feature": "F27.browser-screenshot",+ "entry": "OpenHands/OpenHands#17383 — [Bug]: Browser tooling fails due to incompatible MCP dependency version",+ "result": "blocked",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "Actual pinned runtime browser-use 0.11.13/mcp 1.30.0/fastmcp 3.4.8 BrowserToolExecutor constructs. The report’s supported Canvas/global browser operation still needs a genuine browser-enabled agent launch and authorized LLM to issue a browser_* operation; a constructor probe alone does not establish that user flow.",+ "artifacts": [+ "evidence/F27.browser-screenshot/dependency-17383-5152.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.778Z",+ "feature": "F20.app-backend-view",+ "entry": "OpenHands/software-agent-sdk#5648 — [Bug]: /app-backends WebSocket bridge negotiates no subprotocol (expected the backend's choice, qa.v1)",+ "result": "blocked",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas.",+ "artifacts": [+ "evidence/F20.app-backend-view/bridge-unavailable.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.882Z",+ "feature": "F20.app-backend-view",+ "entry": "OpenHands/software-agent-sdk#5646 — [Bug]: Canvas App backend stays 'unhealthy' after one slow health probe instead of returning to 'ready'",+ "result": "blocked",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas.",+ "artifacts": [+ "evidence/F20.app-backend-view/bridge-unavailable.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:48.987Z",+ "feature": "F20.app-backend-view",+ "entry": "OpenHands/software-agent-sdk#5637 — [Bug]: Canvas backend outlives a SIGKILLed agent-server; restart reports it stopped and starts a second one",+ "result": "blocked",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas.",+ "artifacts": [+ "evidence/F20.app-backend-view/bridge-unavailable.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:49.096Z",+ "feature": "F26.remote-backend",+ "entry": "OpenHands/software-agent-sdk#5633 — [Bug]: Failed run blocks the event loop while logging a rich traceback: /alive takes ~5 s, not ~100 ms",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Temporarily replaced only the owned Agent Server with the same pinned interpreter and state under genuine default Rich logging (LOG_JSON/CI/LOG_RICH_TRACEBACKS unset). Canvas Send uses tools=[] and a dummy LLM pointing at closed 127.0.0.1:9; 60 concurrent direct /alive probes had maximum latency 2.091 s while the conversation ended error. No provider reachable. Restored normal launcher and doctor afterward.",+ "artifacts": [+ "evidence/F26.remote-backend/rich-failure-latency-5633.json",+ "evidence/F26.remote-backend/rich-failure-5633.png"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:49.203Z",+ "feature": "F03.launch-in-workspace",+ "entry": "OpenHands/software-agent-sdk#5630 — [Bug]: Sub-agent file naming a missing skill makes the first message return 500 instead of being skipped",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Owned workspace has .agents/agents/qa-bad-skill.md naming nonexistent qa-f27-missing-skill. With only a closed-loopback dummy LLM profile and autotitling disabled, Home launch in that folder returns the missing-skill error and remains on Home. Two genuine message POSTs with run:false and a create with initial_message run:false each returned HTTP 500 before inference.",+ "artifacts": [+ "evidence/F03.create-error-toast/missing-skill-home-5630-2.png",+ "evidence/F03.create-error-toast/bad-skill-5630.aria.txt"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:49.311Z",+ "feature": "F27.browser-screenshot",+ "entry": "OpenHands/software-agent-sdk#5152 — Unlocked installs resolve browser-use 0.11.13 with mcp 2.x — browser tools silently disappear",+ "result": "pass",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "The actual unlocked pinned runtime resolves openhands-tools 1.54.0, browser-use 0.11.13, mcp 1.30.0 and fastmcp 3.4.8. Its genuine BrowserToolExecutor constructs successfully using its own interpreter and private HOME. The reported mcp 2.x/list_tools import failure does not occur in this exact dependency set; no model was called.",+ "artifacts": [+ "evidence/F27.browser-screenshot/dependency-17383-5152.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:49.420Z",+ "feature": "F23.configuration",+ "entry": "OpenHands/automation#557 — Prompt and plugin presets ignore the selected agent profile at run time",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Saved bare qa-backfill-bare agent profile (tools=[], unique persona, closed-loopback qa-offline LLM), created prompt preset then PATCHed its selected agent_profile_id. Real UI Run now created a linked conversation whose launched_agent_profile is null, selected persona is absent and default tools are present; the selected profile was ignored before the deliberate closed-loopback LLM connection failure.",+ "artifacts": [+ "evidence/F23.run-linked-conversation/profile-557.json",+ "evidence/F23.run-linked-conversation/profile-557.aria.txt"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:49.526Z",+ "feature": "F03.cancel-in-flight-run",+ "entry": "OpenHands/automation#514 — fix(local): stop BashCommand when an automation run is cancelled",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Clicked Home owned worker Cancel run; toast Automation run cancelled and Activity Log CANCELLED. The python3 worker.py remains in this run launcher ancestry after cancellation, despite no provider or external worker calls.",+ "artifacts": [+ "evidence/F03.cancel-in-flight-run/owned-worker-post-cancel.json",+ "evidence/F23.activity-log/cancelled-514.aria.txt"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:49.631Z",+ "feature": "F21.card-grid",+ "entry": "OpenHands/automation#347 — Run summaries are only readable one automation at a time, and the SQLite engine's undeclared pool turns the resulting fan-out into 30s stalls + HTTP 500",+ "result": "blocked",+ "expected": "Execute the original reported trigger in the actual current configuration.",+ "actual": "The owned 25-automation local dashboard completed run summaries with HTTP 200 and no 30 s stall on automation 1.19.3. This narrow live probe does not account for the issue’s missing batch-summary API/pool configuration contracts or withheld original lock workload. A failing dashboard recipe needs the original DB load/lock setup and redacted service-log/HAR timing evidence, linked to Canvas #16559.",+ "artifacts": [+ "evidence/F21.card-grid/scale-347.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:49.734Z",+ "feature": "F21.card-grid",+ "entry": "Local dashboard / 25 inert automations — narrow probe for automation#347",+ "result": "pass",+ "expected": "All owned local run-summary requests return 200 without a30s stall.",+ "actual": "All25 summaries returned200; current UI throttle applies. Original withheld lock/workload and independent service contracts remain unverified.",+ "artifacts": [+ "evidence/F21.card-grid/scale-347.json"+ ],+ "note": "This does not resolve the whole backend issue.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T06:17:49.836Z",+ "feature": "F22.setup-tarball-create",+ "entry": "OpenHands/automation#343 — setup_script_path is validated, stored and tested — but never read; only a root setup.sh ever runs",+ "result": "fail",+ "expected": "The reported user flow behaves correctly on the frozen target.",+ "actual": "Uploaded owned tarball scripts/setup.sh writes .qa-setup-marker, main.py requires marker; persisted setup_script_path scripts/setup.sh. UI Run now genuine BashOutput exits1 with QA_NESTED_SETUP_WAS_SKIPPED, proving nested setup ignored.",+ "artifacts": [+ "evidence/F23.activity-log/nested-status-proof.json"+ ],+ "note": "Recorded after fixture restoration; actual action/URL is described in the issue steps and artifact. Frozen fa1f724; no provider reachable in offline checks.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-integrations-worker"+ },+ {+ "ts": "2026-10-12T09:43:15.596Z",+ "feature": "F10.edit",+ "entry": "Bug backfill: OpenHands/OpenHands#17806",+ "result": "pass",+ "expected": "Changing only the model preserves the stored key.",+ "actual": "Two UI saves and actual switches kept the dummy key and edited model; provider success untested.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F10.edit/key-edit-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F10.edit/17806-dummy-key-preserved.png"+ ],+ "note": "Scoped Local pre-model diagnostic on frozen fa1f724 / SDK1.54.0; all model endpoints are refused loopback.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T09:43:18.090Z",+ "feature": "F11.agent-uses-connection",+ "entry": "Bug backfill: OpenHands/OpenHands#17803",+ "result": "fail",+ "expected": "New chats resolve the saved provider connection URL.",+ "actual": "Stored connection /v2 then /v3; both new runtimes retained /v1.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F11.agent-uses-connection/url-edit-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F11.agent-uses-connection/17803-stale-loopback-url.png"+ ],+ "note": "Scoped Local pre-model diagnostic on frozen fa1f724 / SDK1.54.0; all model endpoints are refused loopback.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T09:57:58.500Z",+ "feature": "F10.edit",+ "entry": "Bug backfill: OpenHands/OpenHands#17806",+ "result": "pass",+ "expected": "Model-only Basic and Advanced saves keep the untouched stored key.",+ "actual": "Two Advanced and one protected Basic saves/swaps kept the dummy key and changed model; provider success untested.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F10.edit/key-edit-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F10.edit/17806-dummy-key-preserved.png",+ "backfill-18309-settings-rechecks/evidence/F10.edit/17806-basic-key-preserved.png"+ ],+ "note": "Corrected Basic trial used owned default OpenAI loopback endpoints; one earlier rejected dummy-key provider validation was a setup mistake and is excluded from verdict.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T09:57:59.888Z",+ "feature": "F19.enabled-autoload",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5628",+ "result": "fail",+ "expected": "After plugin disable and restart, its skills/command do not survive resume.",+ "actual": "Two resumed conversations retained both plugin skills and activated the disabled command; hooks stayed dropped. Fresh disabled chat activated neither.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F19.enabled-autoload/resume-plugin-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F19.enabled-autoload/5628-disabled-resume-activation.png"+ ],+ "note": "Pre-model state/activation only; no successful reply or plugin MCP reconciliation claimed.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T10:23:01.095Z",+ "feature": "F10.custom-model",+ "entry": "Bug backfill: OpenHands/OpenHands#17261",+ "result": "fail",+ "expected": "Configured OpenAI-compatible connection dispatches the saved model namespace.",+ "actual": "Both actual starts failed at provider selection for saved qwen despite a configured OpenAI connection.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F10.custom-model/17261-qwen-dispatch.png"+ ],+ "note": "Two bounded current Local diagnostics before successful inference. Exact historical installation/provider/account branches listed in the result ledger.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T10:23:03.758Z",+ "feature": "F10.connection-link",+ "entry": "Bug backfill: OpenHands/OpenHands#17195",+ "result": "fail",+ "expected": "Configured OpenAI-compatible connection dispatches the saved model namespace.",+ "actual": "Both starts rejected the reported ollama-local namespace despite a saved custom_openai connection.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F10.connection-link/17195-ollama-prefix-dispatch.png"+ ],+ "note": "Two bounded current Local diagnostics before successful inference. Exact historical installation/provider/account branches listed in the result ledger.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T10:23:09.194Z",+ "feature": "F13.acp-conversation",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5466",+ "result": "fail",+ "expected": "Codex uses the pinned adapter when a stale binary is on PATH.",+ "actual": "Both actual ACP initializations reported the published stale 0.16.0 adapter rather than the pinned 1.10.0, and the warning misattributed selection to npx fallback.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F13.acp-conversation/stale-adapter-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F13.acp-conversation/5466-stale-published-adapter.png"+ ],+ "note": "Two bounded current Local diagnostics before successful inference. Exact historical installation/provider/account branches listed in the result ledger.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T10:24:19.711Z",+ "feature": "F10.default-used",+ "entry": "Bug backfill: OpenHands/OpenHands#16578",+ "result": "pass",+ "expected": "Current Bedrock named-profile initialization imports boto3.",+ "actual": "Both starts imported boto3 and failed at missing named AWS profile before AWS/provider HTTP.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F10.default-drives-conversation/bedrock-import-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F10.default-drives-conversation/16578-boto3-import-diagnostic.png"+ ],+ "note": "Current Local SDK1.54.0 import diagnostic only; valid AWS credentials/inference and historical Linux/fresh dependency resolution untested.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T10:37:24.184Z",+ "feature": "F10.custom-model",+ "entry": "Bug backfill: OpenHands/OpenHands#17261",+ "result": "fail",+ "expected": "Configured OpenAI-compatible connection dispatches the stored model namespace.",+ "actual": "Both actual qwen starts failed provider selection before HTTP despite a saved OpenAI connection.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F10.custom-model/qwen-dispatch-diagnostic.json"+ ],+ "note": "Sanitized immutable prior error summaries, added after completed teardown; raw stack/event files remain private. Scope is current Local dispatch, not working external inference.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T10:37:27.792Z",+ "feature": "F10.connection-link",+ "entry": "Bug backfill: OpenHands/OpenHands#17195",+ "result": "fail",+ "expected": "Configured OpenAI-compatible connection dispatches the stored model namespace.",+ "actual": "Both actual ollama-local/qwen starts failed provider selection despite a custom_openai connection; exact original suffix omitted.",+ "artifacts": [+ "backfill-18309-settings-rechecks/evidence/F10.connection-link/ollama-prefix-dispatch-diagnostic.json"+ ],+ "note": "Sanitized immutable prior error summaries, added after completed teardown; raw stack/event files remain private. Scope is current Local dispatch, not working external inference.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-settings-rechecks"+ },+ {+ "ts": "2026-10-12T05:36:54.422Z",+ "feature": "F15.save-dirty-state",+ "entry": "Settings > Condenser > All, stale-query reconnect",+ "result": "fail",+ "expected": "Unsaved Max Size=111 remains and Save remains enabled after a changed server baseline refetch.",+ "actual": "Max Size becomes 300 and Save disables on the mounted form, twice; no UI Save/reset/scope change.",+ "artifacts": [+ "evidence/F15.save-dirty-state/16948-before-refetch.png",+ "evidence/F15.save-dirty-state/16948-after-refetch.png",+ "evidence/F15.save-dirty-state/network-state-and-refetch.gif"+ ],+ "note": "OpenHands/OpenHands#16948; TARGET fa1f724ab5737a17453ad6081129eb91c5a423c6; SDK1.54.0. Real offline/online networking, no routes intercepted. Clock and Max Size restored.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T05:49:33.441Z",+ "feature": "F10.create-prefill",+ "entry": "Settings → LLM → Add/Edit with an active custom-base-URL profile",+ "result": "fail",+ "expected": "The embedded editor starts on Basic per issue 17801 acceptance criteria.",+ "actual": "Both Add and Edit selected All twice with QA_LOOPBACK custom base URL and temperature 0.4.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.create-prefill/17801-add-all.png",+ "backfill-18309-settings-worker/evidence/F10.edit/17801-edit-all.png"+ ],+ "note": "Frozen target fa1f724; SDK1.54.0. No model request or Save.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T05:49:35.216Z",+ "feature": "F10.model-picker",+ "entry": "Settings → LLM → Add → Basic → OpenRouter → off-catalog model",+ "result": "fail",+ "expected": "Typing thinkingmachines/inkling-small:free remains after blur and is qualified with selected provider.",+ "actual": "Two fill/Tab cycles returned an empty model. Covers root 17577 and the Basic dead end in 17233; no provider save was attempted.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.model-picker/17577-off-catalog.png"+ ],+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T05:49:36.997Z",+ "feature": "F10.model-picker",+ "entry": "Settings → LLM → Add → Basic → Azure",+ "result": "fail",+ "expected": "Azure Basic offers its required Base URL field.",+ "actual": "Base URL count was 0 on Basic, 1 on Advanced, and 0 again on Basic. Runtime Azure auth path remains untested.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.model-picker/16726-azure-basic.png"+ ],+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T05:56:24.223Z",+ "feature": "F13.set-active",+ "entry": "Settings → Agents → Default row menu",+ "result": "fail",+ "expected": "Consistent Default wording.",+ "actual": "The badge says Default, and the menu says Set as active on two openings.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F13.set-default/17426-wording.png"+ ],+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T06:14:27.624Z",+ "feature": "F10.create-validation",+ "entry": "Settings LLM Add → loopback endpoint →5 retries → Save",+ "result": "fail",+ "expected": "Validation rejects an unreachable provider promptly and prevents saving.",+ "actual": "Two real validations hung until the30s browser timeout; each fell through to a201profileSave. With1retry, both fast checks had rejected with Connection error and404.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved.png",+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved-repeat.png"+ ],+ "note": "Root15778 and SDK5100slow branch. The exhausted-quota429branch needs an approved real provider fixture; no synthetic response used.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T06:14:30.569Z",+ "feature": "F05.llm-profile-picker",+ "entry": "Local Home and conversation →8 saved profiles",+ "result": "pass",+ "expected": "Menu within viewport, first and last rows reachable.",+ "actual": "Home at1512x720,desktop,phone,narrow and conversation at1512x720/phone permitted inner scroll and first/last selection; Cloud and ACP model catalogue untested.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F05.llm-profile-picker/17239-conversation-phone.png",+ "backfill-18309-settings-worker/evidence/F05.model-picker/17239-eight-1512x720.png"+ ],+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:19.785Z",+ "feature": "F18.card",+ "entry": "Bug backfill: OpenHands/OpenHands#18287",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "After filtering a personal/bundled name collision, the summary said one result while two then three stale skill-creator cards remained.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F18.card/18287-ghost-filtered-card.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:22.554Z",+ "feature": "F10.create-prefill",+ "entry": "Bug backfill: OpenHands/OpenHands#17801",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Add and Edit selected All twice for a profile with a custom base URL and temperature 0.4, rather than the requested Basic starting tab.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.create-prefill/17801-add-all.png",+ "backfill-18309-settings-worker/evidence/F10.edit/17801-edit-all.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:26.453Z",+ "feature": "F10.model-picker",+ "entry": "Bug backfill: OpenHands/OpenHands#17577",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Basic OpenRouter discarded the off-catalog model ID on blur in two trials; the model field became empty.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.model-picker/17577-off-catalog.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:28.804Z",+ "feature": "F10.delete",+ "entry": "Bug backfill: OpenHands/OpenHands#17520",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Two deletes returned 200 while title-cleanup PATCH returned 500; the UI kept the deleted row although its API returned 404. Temporary own-file immutable flags were restored immediately and doctor passed.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.delete/17520-immutable-cleanup-error.png",+ "backfill-18309-settings-worker/evidence/F10.delete/17520-immutable-repeat.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:31.482Z",+ "feature": "F13.set-active",+ "entry": "Bug backfill: OpenHands/OpenHands#17426",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "The active agent profile badge says Default while its menu says Set as active, on two openings.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F13.set-default/17426-wording.png",+ "backfill-18309-settings-worker/evidence/F13.set-default/17426-wording.aria.txt"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:34.909Z",+ "feature": "F10.model-picker",+ "entry": "Bug backfill: OpenHands/OpenHands#17233",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Reproduced the Basic entry dead end described by concrete companion #17577. The authenticated Advanced/save branch was not exercised.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.model-picker/17577-off-catalog.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:37.367Z",+ "feature": "F17.server-health",+ "entry": "Bug backfill: OpenHands/OpenHands#16954",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Two SPA switches from a freshly probed Local identity to unprobed QA_SCOPE reused the same Reachable verdict. The identities use ingress/direct URLs of one owned stack; independent credential and delayed-probe branches were not tested.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F17.server-health/16954-spa-cross-scope.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:42.510Z",+ "feature": "F15.save-dirty-state",+ "entry": "Bug backfill: OpenHands/OpenHands#16948",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Reproduced twice on a healthy frozen stack: an unsaved Condenser Max Size=111 was replaced by the changed server baseline=300 and Save disabled after a real stale-query background refetch. No UI Save, reset or scope change occurred. Clock/online state and Max Size=240 restored.",+ "artifacts": [+ "evidence/F15.save-dirty-state/16948-before-refetch.png",+ "evidence/F15.save-dirty-state/16948-after-refetch.png",+ "evidence/F15.save-dirty-state/16948-repeat.png",+ "evidence/F15.save-dirty-state/network-state-and-refetch.mp4",+ "evidence/F15.save-dirty-state/network-state-and-refetch.gif"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:44.376Z",+ "feature": "F18.page",+ "entry": "Bug backfill: OpenHands/OpenHands#16893",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Two SPA backend switches made zero skills requests and omitted a newly arranged card. Reload then fetched actual data and showed it. Registry identities share one physical stack, with real fixture data changes between loads.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F18.page/16893-spa-stale-cache.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:46.350Z",+ "feature": "F19.launch-unicode",+ "entry": "Bug backfill: OpenHands/OpenHands#16735",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "A real UTF-8 base64 launch path corrupted Chinese/emoji parameter intoLatin1mojibake twice, including reload. This tests decoding; no source-builder function evaluation was counted asUIproof.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F19.launch-unicode/16735-unicode-parameters.png",+ "backfill-18309-settings-worker/evidence/F19.launch-unicode/16735-unicode-parameters.aria.txt"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:50.658Z",+ "feature": "F10.model-picker",+ "entry": "Bug backfill: OpenHands/OpenHands#16726",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Azure Basic had no Base URL input; Advanced had one, and returning to Basic hid it again. No authenticated Azure conversation was attempted.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.model-picker/16726-azure-basic.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:53.537Z",+ "feature": "F18.toggle",+ "entry": "Bug backfill: OpenHands/OpenHands#15694",+ "result": "pass",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "The disabled skill stayed off in two eight-second samples and reloads; the server retained its disabled name. The report is intermittent, so longer or different timing remains unverified.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F18.toggle/15694-disabled-persists.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:55.236Z",+ "feature": "F10.create-validation",+ "entry": "Bug backfill: OpenHands/OpenHands#15778",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Fast refused-loopback validation rejected the profile, but two validations with five retries reached the 30-second browser timeout and then saved with 201.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.create-validation/15778-invalid-rejected.png",+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved.png",+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved-repeat.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:57.998Z",+ "feature": "F15.verification-page",+ "entry": "Bug backfill: OpenHands/OpenHands#15811",+ "result": "pass",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Default Basic and Advanced both show two switches. Enabling Confirmation Mode reveals Security Analyzer only on Advanced. The report asks whether this is intentional; mapped as conditional-tier clarification.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F15.verification-page/15811-basic.aria.txt",+ "backfill-18309-settings-worker/evidence/F15.verification-page/15811-advanced.aria.txt",+ "backfill-18309-settings-worker/evidence/F15.verification-critic-fields/15811-conditional-advanced.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:04:59.830Z",+ "feature": "F16.analytics",+ "entry": "Bug backfill: OpenHands/OpenHands#17898",+ "result": "pass",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "The hard-DNT build sent zero PostHog requests before and after opt-in/Save/reload, and created no PostHog storage key. Consent was restored off.",+ "artifacts": [],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:02.380Z",+ "feature": "F19.launch-trust",+ "entry": "Bug backfill: OpenHands/OpenHands#17162",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Crafted evil.example/github.com/...source renders onlyAll-Hands-AI/openhands-plugin and omits evil.example twice; Start remained disabled and no plugin was fetched.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F19.launch-trust/17162-source-spoof.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:03.950Z",+ "feature": "F15.save-errors",+ "entry": "Bug backfill: OpenHands/OpenHands#15813",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Max Size accepted -5 and enabled Save twice. The server rejected the save and kept 240; the client input floor is missing.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F15.save-errors/15813-negative-enabled.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:11.188Z",+ "feature": "F19.enabled-autoload",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5718",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Two malformed-manifest first sends showed Invalid JSON and Sending, persisted no events and stayed idle. A malformed skill under a valid manifest instead initialized successfully.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F19.enabled-autoload/5718-broken-manifest-first-send.png",+ "backfill-18309-settings-worker/evidence/F19.enabled-autoload/5718-broken-manifest-repeat.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:14.920Z",+ "feature": "F19.add-plugin",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5645",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Two Add attempts with an unknown ref on a warm local Git cache succeeded and installed cached version 1.0.0 instead of rejecting the missing ref.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F19.add-plugin/5645-unknown-ref.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:18.974Z",+ "feature": "F17.custom-add",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5643",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "An empty MCP body returned 201 twice and persisted enabled:true; real UI-started initialization lost the healthy echo tool. The invalid entry had no UI card.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F17.runtime-tools/5643-mixed-empty-init.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:21.481Z",+ "feature": "F11.list",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5632",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "The own provider-store lock delayed unrelated /alive responses by 25.416 and 28.060 seconds; the second list request returned 503 after 30.052 seconds. Locks were released and doctor passed.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F11.list/5632-lock-blocks-ui.png",+ "backfill-18309-settings-worker/evidence/F11.list/store-lock-diagnostic.json"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:23.159Z",+ "feature": "F11.create-validation",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5624",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Two real Add forms accepted the redacted placeholder and persisted keyless connections. Typing it in Edit changed key-set to true. Keyboard Save was used; whitespace-only and provider-auth branches were not claimed.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F11.create-validation/5624-sentinel-created.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:30.000Z",+ "feature": "F10.create-validation",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5100",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "The slow branch reproduced twice: five-retry loopback validation lasted until the 30-second client abort, then Canvas saved the profile. The separate exhausted-quota 429 branch requires an approved real endpoint.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved.png",+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved-repeat.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:32.229Z",+ "feature": "F17.agent-uses-server",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#4891",+ "result": "pass",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Missing stdio and unreachable HTTP MCP sources both allowed initialization and preserved the healthy sibling tool, then reached the expected loopback model error. No successful reply was claimed.",+ "artifacts": [],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:34.176Z",+ "feature": "F15.agent-context-effect",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#4709",+ "result": "pass",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Canvas launches used fresh runtime datetimes despite naturally stale settings and an explicit old-date fixture. The old date was absent from the initialized prompt.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F15.agent-context-page/datetime-diagnostic.json"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:35.895Z",+ "feature": "F13.acp-credentials",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#4629",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "An empty run-owned Gemini OAuth file produced the already-signed-in banner on two fresh ACP forms. Provider auth priority and version migration were not exercised.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F13.acp-credentials/4629-empty-file-signed-in.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:38.550Z",+ "feature": "F13.active-drives-conversation",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#4533",+ "result": "pass",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Two seeded-default UI launches preserved the dummy key in runtime state and reached loopback connection refusal. This verifies key propagation, not successful provider authentication.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F13.active-drives-conversation/default-key-diagnostic.json"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:41.710Z",+ "feature": "F19.refresh",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#4363",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Completed Update changed persisted requested_ref from v1 to null. Visible version stayed 1.0.0 in this cache, so a default-HEAD jump was not observed.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F19.refresh/4363-pinned-before.aria.txt",+ "backfill-18309-settings-worker/evidence/F19.refresh/pin-loss-diagnostic.json"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:44.321Z",+ "feature": "F18.personal-skill-dirs",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#4252",+ "result": "pass",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "A persistence-folder skill appeared in the UI and in available/activated skills in two actual initialized chats. This verifies discovery before the model reply, not historical Docker mounts.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F18.personal-skill-dirs/4252-legacy-skill-loaded.png"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:47.039Z",+ "feature": "F17.agent-uses-server",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#4246",+ "result": "pass",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "Two actual stdio timeouts left initialization after about 30 seconds, persisted the prompt/user message and reached loopback model error. This settles the current Local initialization loop only, not successful replies or historical Cloud/SSE behavior; sibling tools were lost.",+ "artifacts": [],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:49.114Z",+ "feature": "F14.create",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5622",+ "result": "fail",+ "expected": "The linked issue's requested behavior holds on its current Canvas consumer.",+ "actual": "With an approved replacement-cipher/old-encrypted-state fixture, secret creation rewrote an old entry and its dummy getter returned 404. Two unrelated settings saves cleared the old LLM key. Original cipher and stores were restored immediately; restart and doctor passed.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F14.create/5622-before-cipher-change.png",+ "backfill-18309-settings-worker/evidence/F14.create/5622-after-new-secret.png",+ "backfill-18309-settings-worker/evidence/F14.create/5622-settings-save.png",+ "backfill-18309-settings-worker/evidence/F14.create/cipher-mismatch-diagnostic.json"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:05:50.771Z",+ "feature": "F13.openhands-options",+ "entry": "Bug backfill: OpenHands/software-agent-sdk#5519",+ "result": "blocked",+ "expected": "Complete the issue's affected user path once its stated prerequisite is available.",+ "actual": "A complete scoped worker turn needs an approved working model and the affected legacy profile to dispatch a child and invoke TaskTrackingTool. Current Canvas Save added task_tracker despite its unchecked box, and real parent initialization exposed task/task_tracker; this is not a child-turn verdict.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F13.openhands-options/scoped-task-diagnostic.json"+ ],+ "note": "Retrospective import of captured real live checks on frozen fa1f724 / SDK1.54.0. Exact actual page, setup and scope are in the owning map recipe; recorded URL reflects final cleanup.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T08:09:40.611Z",+ "feature": "F15.save-dirty-state",+ "entry": "Bug backfill: OpenHands/OpenHands#16948",+ "result": "fail",+ "expected": "A background refetch keeps the unsaved draft.",+ "actual": "Two real stale-query/refetch trials replaced draft111 with300 and disabledSave; original240/network/clock restored.",+ "artifacts": [+ "backfill-18309-settings-worker/evidence/F15.save-dirty-state/16948-before-refetch.png",+ "backfill-18309-settings-worker/evidence/F15.save-dirty-state/16948-after-refetch.png",+ "backfill-18309-settings-worker/evidence/F15.save-dirty-state/network-state-and-refetch.mp4"+ ],+ "note": "Recorded earlier during the real run; this corrects only relative artifact paths in its retrospective import.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-settings-worker"+ },+ {+ "ts": "2026-10-12T04:56:43.410Z",+ "feature": "F14.list",+ "entry": "Settings > Secrets",+ "result": "pass",+ "expected": "Secrets list or empty-state visible",+ "actual": "Secrets page opened on the healthy isolated target",+ "artifacts": [+ "evidence/F14.list/bootstrap-secrets.png"+ ],+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface"+ },+ {+ "ts": "2026-10-12T10:26:58.926Z",+ "feature": "F06.reconnect",+ "entry": "Conversation > durable messages > same-page backend reconnect (OpenHands/OpenHands#17619)",+ "result": "pass",+ "expected": "Scoped durable-event cursor check: reconnect uses the latest held sequence (REST count minus one), retaining the same page; a recovery marker appears once.",+ "actual": "The initial session socket used after_seq=-1. Three genuine run:false user messages yielded events/count=7; the same open page reconnected after an owned backend restart with after_seq=6. A recovery message appeared once. A recorded repeat advanced -1 to2 with count3. No frame bodies were captured, so this does not quantify server replayed payloads or cost.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/socket-cursor-17619.json",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/socket-cursor-video-17619.json",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/cursor-recovery-17619.mp4",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/recovered-video-17619.png"+ ],+ "note": "Recorded after the live checks and teardown; check date2026-10-12, frozen product fa1f724ab5737a17453ad6081129eb91c5a423c6, SDK1.54.0, Chromium153.0.8010.12. The owned run remains stoppedAt 2026-10-12T10:10:24.208Z; no service was restarted to add this entry. Scoped pass only for current after_seq subscription cursor/recovery. No frame bodies were recorded, so zero payload replay/server cost, provider streaming, former timestamp protocol and Cloud are unverified; idle UI outage labels are not graded.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "ts": "2026-10-12T10:26:58.993Z",+ "feature": "F06.stop-resume",+ "entry": "Composer > held dummy request > Stop > controlled crash recovery (OpenHands/software-agent-sdk#5626)",+ "result": "pass",+ "expected": "Scoped held-request pause durability: Canvas Stop immediately saves paused in GET and base_state.json; paused survives a controlled exact-owned-PID crash and restart.",+ "actual": "Actual Canvas Stop displayed Stopped; GET and base_state.json both immediately reported paused while a dummy loopback model request was held. The exact launcher-owned Agent Server PID was SIGKILLed, then the own run restarted; GET, disk and reloaded UI retained paused. The accept-only fixture received one connection, generated no response, and bounded-exited cleanly.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/bounded-inflight-5626.png",+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/pause-durability-5626.json",+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/owned-crash-5626.json",+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/pause-after-crash-5626.json",+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/paused-after-crash-5626.png"+ ],+ "note": "Recorded after the live checks and teardown; check date2026-10-12, frozen product fa1f724ab5737a17453ad6081129eb91c5a423c6, SDK1.54.0, Chromium153.0.8010.12. The owned run remains stoppedAt 2026-10-12T10:10:24.208Z; no service was restarted to add this entry. Scoped pass only for the held dummy-loopback request pause/crash variant. The accept-only endpoint generated no model response. Successful model/tool Play/resumption and other platforms are unverified.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "ts": "2026-10-12T10:26:59.060Z",+ "feature": "F06.reconnect",+ "entry": "Conversation > sandbox pause > new tab comparison > deletion (OpenHands/software-agent-sdk#5635)",+ "result": "fail",+ "expected": "Sandbox pause and deletion close original subscribed sockets; after pause the original Canvas tab receives new durable messages or reconnects to receive them.",+ "actual": "prepare-for-sandbox-pause returned204 but the original communicating session socket stayed open for over10s. A newly opened Canvas tab received a new run:false message; the old tab did not. DELETE returned200 and GET404, but both held sockets stayed open over10s. Canvas later redirected through metadata polling; no claim of permanently stale navigation is made.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-delete-sockets-5635.json",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-new-tab-5635.png",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-stale-tab-5635.png",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/deleted-stale-tab-5635.png"+ ],+ "note": "Recorded after the live checks and teardown; check date2026-10-12, frozen product fa1f724ab5737a17453ad6081129eb91c5a423c6, SDK1.54.0, Chromium153.0.8010.12. The owned run remains stoppedAt 2026-10-12T10:10:24.208Z; no service was restarted to add this entry. Canvas later redirected through metadata polling; this is distinct from timely socket closure. Socket metadata only, no frame bodies/auth data.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "ts": "2026-10-12T10:26:59.128Z",+ "feature": "F07.download-zip",+ "entry": "Conversation header > Download conversation data during real overlapping receivers (OpenHands/software-agent-sdk#5539)",+ "result": "fail",+ "expected": "Every overlapping trajectory request for one idle conversation completes with a valid ZIP; deliberate receiver aborts are unverified rather than counted as corrupted.",+ "actual": "A 20-second three-client real loopback overlapping receiver made929 attempts:31 completed corrupt200 ZIPs,106HTTP500,269ingress502,133RemoteProtocolError,388 validZIPs and2 intentionally deadline-aborted reads. Both ambiguous Content-Length differences were excluded from the public claim. Actual Canvas Download during overlap produced a ZIP with inspectable entries; that particular UI download is not claimed corrupt. A separate sequential2000-request pass returned validZIPs throughout.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F07.download-zip/trajectory-overlap-5539-summary.json",+ "backfill-18309-surface-rechecks/evidence/F07.download-zip/trajectory-summary-correction.json",+ "backfill-18309-surface-rechecks/evidence/F07.download-zip/overlapping-download-5539.png"+ ],+ "note": "Recorded after the live checks and teardown; check date2026-10-12, frozen product fa1f724ab5737a17453ad6081129eb91c5a423c6, SDK1.54.0, Chromium153.0.8010.12. The owned run remains stoppedAt 2026-10-12T10:10:24.208Z; no service was restarted to add this entry. The actual Canvas download was inspectable and is not individually claimed corrupt. Two ambiguous Content-Length differences were excluded because they could be intentional deadline aborts; original private artifact and correction are retained.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "ts": "2026-10-12T10:26:59.197Z",+ "feature": "F07.rename-inline",+ "entry": "Conversation title > rename with owned metadata write denied (OpenHands/software-agent-sdk#4800)",+ "result": "fail",+ "expected": "A refused metadata write leaves both GET and persisted title unchanged; shutdown must not persist that rejected mutation.",+ "actual": "After a successful baseline UI rename, chmod0500 on only the owned conversation directory made the next actual UI rename fail with PermissionError/HTTP500. GET nevertheless reported the rejected title while meta.json retained the baseline. Permissions were restored; graceful restart then persisted the rejected in-memory title. Tags/autotitle branches were not tested.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-before-4800.json",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-denied-4800.json",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/refused-title-4800.png",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-after-graceful-restart-4800.json"+ ],+ "note": "Recorded after the live checks and teardown; check date2026-10-12, frozen product fa1f724ab5737a17453ad6081129eb91c5a423c6, SDK1.54.0, Chromium153.0.8010.12. The owned run remains stoppedAt 2026-10-12T10:10:24.208Z; no service was restarted to add this entry. Rename branch only; tags/autotitle and ownership-conflict branches unverified. Only the owned directory mode was temporarily changed and restored.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "ts": "2026-10-12T10:26:59.264Z",+ "feature": "F07.rename-inline",+ "entry": "Shared-storage secondary first load > connected Canvas title (OpenHands/software-agent-sdk#5629)",+ "result": "fail",+ "expected": "B first load retains the newer metadata title saved by A after B startup; Canvas displays the newer title after ownership transfer.",+ "actual": "B was started with shared storage before A renamed an idle conversation from BEFORE_B to AFTER_B. A PATCH and disk confirmed AFTER_B. After own A graceful stop, native B events/count returned200 and rewrote disk title to its BEFORE_B startup copy. Real Add Backend→B card showed BEFORE_B in Canvas. The native first-load request is explicit; an earlier sandbox-pause variant did not reach a valid first load and is not the claimed reproduction.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-repeat-before-b-load-5629.json",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-repeat-after-b-load-5629.json",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/shared-first-load-reverted-5629.png"+ ],+ "note": "Recorded after the live checks and teardown; check date2026-10-12, frozen product fa1f724ab5737a17453ad6081129eb91c5a423c6, SDK1.54.0, Chromium153.0.8010.12. The owned run remains stoppedAt 2026-10-12T10:10:24.208Z; no service was restarted to add this entry. The native events/count first-load request is explicit. The earlier sandbox-pause attempt did not reach a valid first load and is not the claimed reproduction.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "ts": "2026-10-12T10:26:59.332Z",+ "feature": "F08.vscode-link",+ "entry": "Cloud editor URL scheme validation; local card diagnostic (OpenHands/OpenHands#16871)",+ "result": "blocked",+ "expected": "Current Cloud editor exposed_urls path rejects unsafe schemes before opening; diagnostic local card menu and drawer behavior are recorded separately.",+ "actual": "The local card menu had no Download via VS Code action; the sole production caller omits showOptions. Current drawer source opens with noopener,noreferrer. The current Cloud exposed_urls transformer scheme path was not driven.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F04.download/vscode-card-action-absent-16871.png",+ "backfill-18309-surface-rechecks/evidence/F04.download/vscode-card-action-absent-16871.aria.txt"+ ],+ "note": "Recorded after the live checks and teardown; check date2026-10-12, frozen product fa1f724ab5737a17453ad6081129eb91c5a423c6, SDK1.54.0, Chromium153.0.8010.12. The owned run remains stoppedAt 2026-10-12T10:10:24.208Z; no service was restarted to add this entry. Diagnostic only for local-menu/source observations. Missing prerequisite: an authorized Cloud account/sandbox with a controlled editor exposed_urls value. No opener exploit or Cloud scheme acceptance outcome was reproduced.",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "ts": "2026-10-12T05:01:02.034Z",+ "feature": "F05.attach-files",+ "entry": "Home > attach > remove > pick same file",+ "result": "fail",+ "expected": "The removed file attaches again",+ "actual": "Re-pick leaves attachment count zero",+ "artifacts": [+ "evidence/F05.attach-files/issue-17471-repick.png"+ ],+ "note": "#17471, frozen fa1f724; dummy loopback LLM only",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:04:54.879Z",+ "feature": "F07.rename-inline",+ "entry": "Chat header > Rename > space > Enter",+ "result": "fail",+ "expected": "Restore previous title without saving",+ "actual": "PATCH title empty; server 422 rejects and restores previous title",+ "artifacts": [+ "evidence/F07.rename-inline/issue-17061-titled.png"+ ],+ "note": "#17061 empty submission reproduced; persistent blank title not observed",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:04:54.986Z",+ "feature": "F02.sidebar-collapse",+ "entry": "Sidebar collapse",+ "result": "fail",+ "expected": "Collapsed rail offers access to existing conversations",+ "actual": "No conversation panel or chooser; expanding required",+ "artifacts": [+ "evidence/F02.sidebar-collapse/issue-15662-collapsed.png"+ ],+ "note": "#15662",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:04:55.090Z",+ "feature": "F04.card-live-status",+ "entry": "Folder + > new empty conversation",+ "result": "fail",+ "expected": "Idle empty conversation uses neutral status",+ "actual": "Idle status uses green conversation-status-active dot; no pulse animation",+ "artifacts": [+ "evidence/F04.card-live-status/issue-16752-empty.png"+ ],+ "note": "#16752 narrower result: green active dot, not flashing",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:08:39.131Z",+ "feature": "F04.pin",+ "entry": "Load more > pin page-two conversation > reload",+ "result": "fail",+ "expected": "Pin survives first-page reload",+ "actual": "Pinned section count becomes zero while conversation still exists",+ "artifacts": [+ "evidence/F04.pin/issue-16399-reload.png"+ ],+ "note": "#16399, 23 empty conversations",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:11:38.054Z",+ "feature": "F04.grouped-folders",+ "entry": "View > By workspace > fresh browser > By workspace",+ "result": "fail",+ "expected": "Same server workspaces group the same conversations",+ "actual": "Existing conversations move to No workspace; registered folder remains empty",+ "artifacts": [+ "evidence/F04.grouped-folders/issue-17386-before-reset.png",+ "evidence/F04.grouped-folders/issue-17386-after-reset.png"+ ],+ "note": "#17386; browser reset retains server workspace and conversations",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:13:45.571Z",+ "feature": "F08.files-tree",+ "entry": "Files > Refresh with 1800 long-path files",+ "result": "fail",+ "expected": "All files below 2000 cap remain listed",+ "actual": "README.md disappears and partial first path creates spurious bb folder",+ "artifacts": [+ "evidence/F08.files-tree/sdk-5638-settled.png"+ ],+ "note": "OpenHands/software-agent-sdk#5638; listing larger than one MiB; real files, no model",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:13:45.970Z",+ "feature": "F08.uncommitted-changes",+ "entry": "Commits with modified tracked my notes.md",+ "result": "fail",+ "expected": "Changes list real filename and every other change",+ "actual": "Scoped git/changes returns 400 and hides Uncommitted changes",+ "artifacts": [+ "evidence/F08.uncommitted-changes/sdk-5642-settled.png"+ ],+ "note": "OpenHands/software-agent-sdk#5642; space-name case only tested",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:19:03.849Z",+ "feature": "F08.uncommitted-changes",+ "entry": "Commits > plain parent workspace with dirty proj repository",+ "result": "fail",+ "expected": "Nested repository changes remain visible",+ "actual": "No commits yet with scoped changes200 empty while child has tracked and untracked changes",+ "artifacts": [+ "evidence/F08.uncommitted-changes/sdk-5641-settled.png"+ ],+ "note": "OpenHands/software-agent-sdk#5641; arranged parent as non-repo by removing run-owned auto-created empty .git",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:19:05.275Z",+ "feature": "F05.profile-identity",+ "entry": "Open failed conversation after profile model changed and fresh browser cleared metadata",+ "result": "fail",+ "expected": "Useful existing model or active profile label",+ "actual": "Select a model despite loaded active QA_surface profile and persisted model openai/gpt-4o",+ "artifacts": [+ "evidence/F05.profile-identity/issue-16263-error.png"+ ],+ "note": "#16263; conversation status error on closed loopback endpoint, no paid model calls",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "ts": "2026-10-12T05:21:35.143Z",+ "feature": "F08.files-rich-plain",+ "entry": "Files > QA_rich.md > Rich",+ "result": "fail",+ "expected": "One compact code block container",+ "actual": "Code fence renders nested dark/gray containers with repeated padding",+ "artifacts": [+ "evidence/F08.files-rich-plain/issue-16802-codeblock.png"+ ],+ "note": "#16802; layout-only inspection of a fenced Python block",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local agent-server 1.54.0",+ "viewport": "1440x1000",+ "run_id": "backfill-18309-surface-worker"+ }+ ]+}diff --git a/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.json b/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.jsonnew file mode 100644index 0000000000..4e439a7d4f--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.json@@ -0,0 +1,8400 @@+{+ "schema_version": 1,+ "issue": "https://github.com/OpenHands/OpenHands/issues/18309",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "maintenance_baseline": "553d4519113192a80fb18376fdb6c1e39fa06387",+ "baseline_advanced": false,+ "inventory_snapshots": {+ "OpenHands/OpenHands": {+ "captured_at": "2026-10-12T04:49:30.813663+00:00",+ "open_bugs": 208+ },+ "OpenHands/software-agent-sdk": {+ "captured_at": "2026-10-12T04:50:37.186198+00:00",+ "open_bugs": 141+ },+ "OpenHands/automation": {+ "captured_at": "2026-10-12T04:50:33.550521+00:00",+ "open_bugs": 13+ }+ },+ "counts": {+ "OpenHands/OpenHands": {+ "mapped": 172,+ "not-reproduced": 10,+ "out-of-scope": 26+ },+ "OpenHands/software-agent-sdk": {+ "mapped": 93,+ "out-of-scope": 41,+ "not-reproduced": 7+ },+ "OpenHands/automation": {+ "mapped": 9,+ "out-of-scope": 4+ }+ },+ "accounting_complete": true,+ "reviewed_evidence_retractions": "Append-only corrections remove mistagged F13.set-default and F10.default-drives-conversation rows; declared F13.set-active and F10.default-used retain the supported checks.",+ "pending_count": 0,+ "unpublished_not_reproduced_comments": 0,+ "runtime": {+ "os": "macOS Darwin 27.0.0 arm64",+ "node": "25.9.0",+ "playwright": "1.63.0",+ "chromium": "153.0.8010.12 (actual HeadlessChrome user agent)",+ "theme_fonts": "app defaults; versions not independently audited",+ "agent": "Codex (GPT-6)",+ "provider_spend": "No real model credentials or paid inference used. Disposable initialization/error fixtures use refused or bounded loopback endpoints; one Basic validation lost its base URL and reached OpenAI with a dummy credential, which was rejected before inference. Fallback URLs were protected before replay.",+ "checkout": "Isolated worktree from frozen main; only this skill changed.",+ "build_commands": [+ "npm ci --ignore-scripts --no-audit --no-fund",+ "npm run build:app"+ ]+ },+ "checks": {+ "completed_at": "2026-10-12 10:33:17 UTC",+ "map_check": "ok",+ "map_coverage": "ok: 31 routes and 29 feature directories accounted for",+ "map_testids": "ok: 1037 cited, 0 unresolved",+ "harness_tests": "57/57 passed (focused Vitest suite)",+ "fixture_tests": "8/8 passed (installed SDK Python with httpx; bounded ZIP/transport and process-group cleanup)",+ "syntax_check": "Node CLI, daemon and WebSocket helper passed; all five Python fixture/test files compile",+ "formatter": "Prettier check passed for all four changed JavaScript files",+ "diff_check": "clean",+ "inventory_audit": "362 unique outcomes, 0 missing, 0 pending, 0 problems",+ "details": {+ "map_check": {+ "ok": true,+ "files": 27,+ "ids": 747,+ "checked": "all",+ "problems": []+ },+ "map_coverage": {+ "ok": true,+ "routes": {+ "total": 31,+ "unmapped": [],+ "excluded": [],+ "fileOnly": []+ },+ "featureComponentDirs": {+ "total": 29,+ "unmapped": [],+ "excluded": [+ "src/components/features/context-menu"+ ]+ }+ },+ "map_testids": {+ "ok": true,+ "cited": 1037,+ "unresolved": [],+ "note": "every cited test id is accounted for in src/"+ }+ },+ "teardown": "All nine owned run records stopped; all owned launcher PIDs absent. Twenty-one recorded ports refuse connections; four were reused by a different launcher outside this worktree with no backfill ownership markers."+ },+ "runs": [+ {+ "runId": "backfill-18309-coordinator",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": false,+ "mode": "local",+ "ports": {+ "ingress": 18820,+ "agentServer": 18821,+ "automation": 18822,+ "frontend": 18823,+ "vscode": 19821+ },+ "startedAt": "2026-10-12T05:31:38.273Z",+ "readyAt": "2026-10-12T05:31:45.877Z",+ "stoppedAt": "2026-10-12T05:34:29.215Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": null,+ "check_counts": {},+ "evidence_files_retained": 0+ },+ {+ "runId": "backfill-18309-coordinator-checks",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": false,+ "mode": "local",+ "ports": {+ "ingress": 18840,+ "agentServer": 18841,+ "automation": 18842,+ "frontend": 18843,+ "vscode": 19841+ },+ "startedAt": "2026-10-12T06:05:36.380Z",+ "readyAt": "2026-10-12T06:05:43.990Z",+ "stoppedAt": "2026-10-12T09:31:53.387Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": "backfill-18309-coordinator-checks/evidence/ledger.jsonl",+ "check_counts": {+ "pass": 1,+ "fail": 3+ },+ "evidence_files_retained": 16+ },+ {+ "runId": "backfill-18309-integrations-worker",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": false,+ "mode": "local",+ "ports": {+ "ingress": 18810,+ "agentServer": 18811,+ "automation": 18812,+ "frontend": 18813,+ "vscode": 19811+ },+ "startedAt": "2026-10-12T05:33:54.431Z",+ "readyAt": "2026-10-12T05:34:02.032Z",+ "restartedAt": "2026-10-12T06:07:44.009Z",+ "stoppedAt": "2026-10-12T06:23:39.105Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": "backfill-18309-integrations-worker/evidence/ledger.jsonl",+ "check_counts": {+ "fail": 20,+ "pass": 7,+ "blocked": 5+ },+ "evidence_files_retained": 66+ },+ {+ "runId": "backfill-18309-local-secondary",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": false,+ "mode": "local",+ "ports": {+ "ingress": 18860,+ "agentServer": 18861,+ "automation": 18862,+ "frontend": 18863,+ "vscode": 19861+ },+ "startedAt": "2026-10-12T09:12:20.264Z",+ "readyAt": "2026-10-12T09:12:26.339Z",+ "stoppedAt": "2026-10-12T09:31:53.555Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": null,+ "check_counts": {},+ "evidence_files_retained": 0+ },+ {+ "runId": "backfill-18309-settings-rechecks",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": false,+ "mode": "local",+ "ports": {+ "ingress": 18820,+ "agentServer": 18821,+ "automation": 18822,+ "frontend": 18823,+ "vscode": 19821+ },+ "startedAt": "2026-10-12T09:31:36.011Z",+ "readyAt": "2026-10-12T09:31:43.622Z",+ "restartedAt": "2026-10-12T10:28:55.029Z",+ "stoppedAt": "2026-10-12T10:33:01.788Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": "backfill-18309-settings-rechecks/evidence/ledger.jsonl",+ "check_counts": {+ "pass": 3,+ "fail": 7+ },+ "evidence_files_retained": 21+ },+ {+ "runId": "backfill-18309-settings-worker",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": false,+ "mode": "local",+ "ports": {+ "ingress": 18800,+ "agentServer": 18801,+ "automation": 18802,+ "frontend": 18803,+ "vscode": 19801+ },+ "startedAt": "2026-10-12T05:28:25.778Z",+ "readyAt": "2026-10-12T05:28:37.880Z",+ "restartedAt": "2026-10-12T07:41:59.289Z",+ "stoppedAt": "2026-10-12T08:05:56.859Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": "backfill-18309-settings-worker/evidence/ledger.jsonl",+ "check_counts": {+ "fail": 30,+ "pass": 9,+ "blocked": 1+ },+ "evidence_files_retained": 78+ },+ {+ "runId": "backfill-18309-surface",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": true,+ "mode": "local",+ "ports": {+ "ingress": 18800,+ "agentServer": 18801,+ "automation": 18802,+ "frontend": 18803,+ "vscode": 19801+ },+ "startedAt": "2026-10-12T04:54:13.393Z",+ "readyAt": "2026-10-12T04:54:36.021Z",+ "stoppedAt": "2026-10-12T04:57:49.620Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": "backfill-18309-surface/evidence/ledger.jsonl",+ "check_counts": {+ "pass": 1+ },+ "evidence_files_retained": 2+ },+ {+ "runId": "backfill-18309-surface-rechecks",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": false,+ "mode": "local",+ "ports": {+ "ingress": 18810,+ "agentServer": 18811,+ "automation": 18812,+ "frontend": 18813,+ "vscode": 19811+ },+ "startedAt": "2026-10-12T09:12:56.383Z",+ "readyAt": "2026-10-12T09:13:03.992Z",+ "restartedAt": "2026-10-12T09:55:55.486Z",+ "stoppedAt": "2026-10-12T10:10:24.208Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": "backfill-18309-surface-rechecks/evidence/ledger.jsonl",+ "check_counts": {+ "pass": 2,+ "fail": 4,+ "blocked": 1+ },+ "evidence_files_retained": 47+ },+ {+ "runId": "backfill-18309-surface-worker",+ "revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "buildId": "d465c70f971b4d8e",+ "builtThisRun": false,+ "mode": "local",+ "ports": {+ "ingress": 18800,+ "agentServer": 18801,+ "automation": 18802,+ "frontend": 18803,+ "vscode": 19801+ },+ "startedAt": "2026-10-12T04:58:12.826Z",+ "readyAt": "2026-10-12T04:58:21.917Z",+ "stoppedAt": "2026-10-12T05:21:42.086Z",+ "pins": {+ "agentServer": "1.54.0",+ "agentCanvas": "1.26.0",+ "automation": "1.19.3"+ },+ "versions": {+ "agentServer": "1.54.0",+ "sdk": "1.54.0"+ },+ "retained_ledger": "backfill-18309-surface-worker/evidence/ledger.jsonl",+ "check_counts": {+ "fail": 11+ },+ "evidence_files_retained": 24+ }+ ],+ "issues": [+ {+ "repository": "OpenHands/OpenHands",+ "number": 18303,+ "title": "[Bug]: \"Subscription models are not loaded yet.\" is shown untranslated",+ "url": "https://github.com/OpenHands/OpenHands/issues/18303",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.subscription"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18302,+ "title": "[Bug]: Onboarding's LLM step overwrites an existing LLM profile that has the model-derived name",+ "url": "https://github.com/OpenHands/OpenHands/issues/18302",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F01",+ "feature_ids": [+ "F01.onboarding-repeat-endpoint"+ ],+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18301,+ "title": "[Bug]: Preserve Automation KV encryption across session-key rotation",+ "url": "https://github.com/OpenHands/OpenHands/issues/18301",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.session-key-rotated",+ "F26.session-key-persist"+ ],+ "basis": "live-reproduced",+ "reason": "Wrote and read an owned Automation KV marker (HTTP 201/200), then control-openhands restart --rotate-key with no AUTOMATION_KV_SECRET override. Fresh-auth KV GET now returns HTTP 500 instead of the prior marker.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **KV survives authentication rotation (`F26.session-key-rotated`, `F26.session-key-persist`).** With no explicit `AUTOMATION_KV_SECRET`, create an owned inert automation and PUT a harmless marker to `/api/automation/v1/kv/qa-session-rotation?automation_id=<id>` through `api --write`; GET must return that marker. Run `restart --rotate-key` and `doctor`, then GET again using the helper's fresh private key. Expected: rotating authentication preserves encrypted automation state. Known failure, reproduced 2026-10-12 against automation 1.19.3: the pre-rotation read is 200 and the fresh-auth read is 500 `Failed to decrypt value` ([#18301](https://github.com/OpenHands/OpenHands/issues/18301)). This uses only disposable state and does not establish the original incident's data-loss history."+ ],+ "artifacts": [+ "evidence/F26.session-key-rotated/kv-rotation-18301.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18299,+ "title": "[Bug]: migration 029 errors with `duplicate column name: consecutive_auth_failures` on SQLite",+ "url": "https://github.com/OpenHands/OpenHands/issues/18299",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "live-reproduced",+ "reason": "Stopped only the owned automation service, backed up its DB, reset alembic_version from 032 to 028 while consecutive_auth_failures exists, restarted the real launcher. Migration duplicate-column signature and automation unavailability occur while launcher stays alive. Private original DB restored afterward.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Partial SQLite migration startup (`F26.runtime-services`).** Stop only the owned automation service, make a cold backup of `<run>/private/automation/automations.db`, and confirm its `automation_git_sync_org_config` already has `consecutive_auth_failures`. In that disposable DB only, set `alembic_version.version_num` to `028`; restart with a bounded timeout, then open `/automations`. Expected: required-service startup fails explicitly or the partial schema recovers without corrupting state. Known failure, reproduced 2026-10-12 against automation 1.19.3: migration 029 reports the duplicate column and exits 3, the launcher remains alive, and Canvas shows **Automations Unavailable** ([#18299](https://github.com/OpenHands/OpenHands/issues/18299)). Restore the cold backup, restart and doctor before continuing. Never apply this fixture to an operator database."+ ],+ "artifacts": [+ "evidence/F26.runtime-services/partial-schema-18299.json",+ "evidence/F26.runtime-services/partial-schema-18299.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18287,+ "title": "[Bug]: Repeated cards for the same skill source when local and bundled skill names collide",+ "url": "https://github.com/OpenHands/OpenHands/issues/18287",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F18",+ "feature_ids": [+ "F18.card",+ "F18.facets"+ ],+ "basis": "live-reproduced",+ "reason": "After filtering a personal/bundled name collision, the summary said one result while two then three stale skill-creator cards remained.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "After filtering a personal/bundled name collision, the summary said one result while two then three stale skill-creator cards remained.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F18.card/18287-ghost-filtered-card.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18234,+ "title": "[Bug]: Pinned automation card on Home hides the run's task summary when the run has a conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/18234",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.home-run-outcome-health"+ ],+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18233,+ "title": "[Bug]: Escape on the closed backend selector clears it: the backend name becomes a placeholder and the status dot disappears",+ "url": "https://github.com/OpenHands/OpenHands/issues/18233",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.selector-dropdown"+ ],+ "basis": "existing-note",+ "reason": "Historical reproduced note dated 2026-10-09 already describes closed-menu Escape clearing the active backend display; source Dropdown and backend selector still own the path.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18210,+ "title": "[Bug]: Home shows dead automation link and manifest errors when automation interface is absent",+ "url": "https://github.com/OpenHands/OpenHands/issues/18210",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F21",+ "feature_ids": [+ "F21.manifest-404"+ ],+ "basis": "existing-note",+ "reason": "Existing missing-manifest recipe and note describe the Home link/toasts; preserved source revision and prior evidence date.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18161,+ "title": "[Bug]: Slack MCP credential probe reports any failed tool call (e.g. \"fetch failed\") as \"Credential check failed\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/18161",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F17",+ "feature_ids": [+ "F17.credential-probe"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18160,+ "title": "[Bug]: Recovery gate (stored backend answers 503) shows two generic \"An error occurred\" toasts from the free-models hydrator",+ "url": "https://github.com/OpenHands/OpenHands/issues/18160",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.frontend-only-returning"+ ],+ "basis": "existing-note",+ "reason": "Existing reproduced 2026-10-08 recovery-gate hydrator/toast note preserved.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18155,+ "title": "[Bug]: Missing-runtime conversations shown as Archived but never hidden, chipped, or unarchivable",+ "url": "https://github.com/OpenHands/OpenHands/issues/18155",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.show-archived-unarchive"+ ],+ "basis": "known-issue",+ "reason": "A Cloud sandbox or local runtime that becomes MISSING and non-resumable; ordinary local pause remains resumable.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18136,+ "title": "[Bug]: Connect Repo on a cloud conversation does not save the repository on the server",+ "url": "https://github.com/OpenHands/OpenHands/issues/18136",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.cloud-only"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Cloud sandbox/repository provider and a second fresh browser to compare server repository metadata.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18117,+ "title": "[Bug]: Inline thinking leaks into the agent chat bubble",+ "url": "https://github.com/OpenHands/OpenHands/issues/18117",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.thinking"+ ],+ "basis": "known-issue",+ "reason": "A live model that emits a leading inline think block on a tool-call turn, plus its finalized ActionEvent.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18108,+ "title": "[Bug]: Files tab shows \"No files in workspace\" on cloud when the agent writes outside /workspace/project",+ "url": "https://github.com/OpenHands/OpenHands/issues/18108",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-tree"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Enterprise Cloud sandbox, no repository selected, with files outside its default working directory.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18063,+ "title": "[Bug]: Composer \"More input actions\" Model submenu runs off the right and bottom edges at narrow widths (third profile and LLM Profiles link can't be clicked or tapped)",+ "url": "https://github.com/OpenHands/OpenHands/issues/18063",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.overflow-menu"+ ],+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18062,+ "title": "[Bug]: Automations: a deleted automation stays on the dashboard, home page and detail page after Run now fails with \"Automation not found\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/18062",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F21.run-now-error",+ "F23.run-now-error"+ ],+ "basis": "existing-note",+ "reason": "Existing dashboard and detail stale-automation notes and recipes preserved; no fresh reproduction claimed.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18029,+ "title": "[Bug]: Locked-to-cloud Agent Canvas shows 'No backend configured' after onboarding despite a healthy configured backend",+ "url": "https://github.com/OpenHands/OpenHands/issues/18029",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.locked-cloud"+ ],+ "basis": "known-issue",+ "reason": "Affected locked-to-cloud deployment, authenticated Cloud session and stale-document/alternate-origin evidence are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17959,+ "title": "[Bug]: Automation templates: a search with no matches shows a blank page, and cards say \"1 MCPs to connect\" for optional integrations",+ "url": "https://github.com/OpenHands/OpenHands/issues/17959",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F22",+ "feature_ids": [+ "F22.templates-search",+ "F22.template-card"+ ],+ "basis": "existing-note",+ "reason": "Both no-result template search and optional-integration pill are already recorded; preserved existing notes.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17955,+ "title": "[Bug]: Add LLM Profile shows the agent's current LLM options (such as temperature) but saves the new profile without them",+ "url": "https://github.com/OpenHands/OpenHands/issues/17955",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.create-prefill"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17953,+ "title": "[Bug]: Backends and Cloud sign-in: Escape closes the whole Manage backends stack, raw 404 toast on shared pages, stray about:blank pop-ups",+ "url": "https://github.com/OpenHands/OpenHands/issues/17953",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.edit-escape",+ "F25.shared-conversation-view",+ "F25.cloud-device-flow"+ ],+ "basis": "existing-note",+ "reason": "All three reported failures already have existing map notes and drive recipes.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17949,+ "title": "[Bug]: Launcher: --allow-lan-session-key is ignored, the port-in-use error prints twice with a stack trace, and backend-only runtime_services lists a frontend",+ "url": "https://github.com/OpenHands/OpenHands/issues/17949",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.host-bind-lan-optin",+ "F26.cli-port-in-use",+ "F26.backend-only"+ ],+ "basis": "existing-note",+ "reason": "All three launcher defects already mapped; retained reproduced 2026-10-08 flag note.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17948,+ "title": "[Bug]: Automation detail and edit: schedule edit silently dropped, wrong run label, slow not-found page, tarball saved as .tar",+ "url": "https://github.com/OpenHands/OpenHands/issues/17948",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.edit-time-cleared",+ "F23.run-no-conversation",+ "F23.not-found",+ "F23.tarball"+ ],+ "basis": "existing-note",+ "reason": "All four automation-detail defects already recorded in Gotchas and corresponding recipes.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17946,+ "title": "[Bug]: Launcher never creates TMUX_TMPDIR, so local Agent Canvas instances share one tmux server and reset each other's terminals",+ "url": "https://github.com/OpenHands/OpenHands/issues/17946",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "existing-note",+ "reason": "Existing npm tmux-isolation known issue preserved; verification launcher intentionally supplies isolation and is not a product-fix claim.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17941,+ "title": "[Bug]: Skill and MCP card toggles can swallow the first click, and keep the new state after a failed save",+ "url": "https://github.com/OpenHands/OpenHands/issues/17941",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F18",+ "feature_ids": [+ "F18.toggle",+ "F18.toggle-error"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17932,+ "title": "[Bug]: Settings save retries a 422 validation error three times before showing it",+ "url": "https://github.com/OpenHands/OpenHands/issues/17932",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F15",+ "feature_ids": [+ "F15.save-errors"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17927,+ "title": "[Bug]: A failed Save on Settings > Application discards the user's unsaved edits",+ "url": "https://github.com/OpenHands/OpenHands/issues/17927",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F16",+ "feature_ids": [+ "F16.save-states"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F16-application-settings.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17922,+ "title": "[Bug]: MCP library lists entries that cannot be installed on a local backend; their install dialog is empty and Install does nothing",+ "url": "https://github.com/OpenHands/OpenHands/issues/17922",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F17",+ "feature_ids": [+ "F17.install-unsupported"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17920,+ "title": "[Bug]: A rejected (Cancel) action leaves its event group spinning as \"k/N actions completed\" after the conversation is idle",+ "url": "https://github.com/OpenHands/OpenHands/issues/17920",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.confirmation-mode"+ ],+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17919,+ "title": "[Bug]: \"Delete all\" conversations deletes only the loaded first page and states a smaller count",+ "url": "https://github.com/OpenHands/OpenHands/issues/17919",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.delete-all"+ ],+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17900,+ "title": "[Bug]: Settings switches are skipped by Tab and invisible to screen readers (SettingsSwitch renders <input hidden>)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17900",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F16",+ "feature_ids": [+ "F16.keyboard"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F16-application-settings.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17899,+ "title": "[Bug]: Git user name and email from Settings > Application are never applied to agent commits",+ "url": "https://github.com/OpenHands/OpenHands/issues/17899",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F16",+ "feature_ids": [+ "F16.git-identity"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F16-application-settings.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17898,+ "title": "[Bug]: VITE_DO_NOT_TRACK=1 still initializes PostHog and loads remote scripts from the telemetry host",+ "url": "https://github.com/OpenHands/OpenHands/issues/17898",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "family": "F16",+ "feature_ids": [+ "F16.analytics"+ ],+ "basis": "live-not-reproduced",+ "reason": "The hard-DNT build sent zero PostHog requests before and after opt-in/Save/reload, and created no PostHog storage key. Consent was restored off.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "passes": 2,+ "steps": [+ "control-openhands browser goto /settings/app",+ "control-openhands browser network --external --filter 'posthog|z.openhands.dev'",+ "control-openhands browser storage",+ "control-openhands browser click 'testid=app-settings-screen >> text=Send anonymous usage data'",+ "control-openhands browser click 'testid=submit-button'",+ "control-openhands browser reload",+ "control-openhands browser network --external --filter 'posthog|z.openhands.dev'",+ "control-openhands browser storage",+ "control-openhands browser click 'testid=app-settings-screen >> text=Send anonymous usage data'",+ "control-openhands browser click 'testid=submit-button'",+ "control-openhands doctor"+ ],+ "evidence": [],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Two network snapshots reported zero PostHog/telemetry-host requests around the real consent-switch Save/reload action; browser storage had no PostHog key. Consent was restored off. The check did not separately record the persisted consent-true field, and covers the hard VITE_DO_NOT_TRACK build only.",+ "replay_setup": {+ "working_directory": "<checkout>",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "build_flag": "VITE_DO_NOT_TRACK=1 at build time (the existing frozen harness build; fingerprint d465c70f971b4d8e); runtime DO_NOT_TRACK=1. A browser-only DNT flag is not equivalent."+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F16-application-settings.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17898#issuecomment-6121789760"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17872,+ "title": "[Bug]: Mock-LLM E2E Tests have failed on every main run since at least 2026-09-22 (five tests fail every run)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17872",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Mock-LLM E2E CI failures and stale test selectors; test harness coverage rather than a proved Canvas user feature failure."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17871,+ "title": "[Bug]: live ACP e2e harness crashes: __EXTENSIONS_SKILLS_DIR__ is not defined in tests/e2e/live-acp/vite-node.config.mts",+ "url": "https://github.com/OpenHands/OpenHands/issues/17871",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Live ACP E2E harness compile-time constant setup; test harness startup rather than the production Canvas path."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17806,+ "title": "[Bug]: Changing LLM model without re-typing the API key wipes the stored key (auth errors on profile switch)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17806",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "family": "F10",+ "feature_ids": [+ "F10.edit"+ ],+ "basis": "live-not-reproduced",+ "reason": "Two Advanced and one protected Basic real model-only saves preserved the untouched hidden dummy key (api_key_set=true); subsequent actual conversation switches materialized the changed model with a populated runtime key. Basic intentionally removed base_url, so the owned server default OPENAI_BASE_URL/OPENAI_API_BASE were set to refused loopback before its successful Save. Scope is current Local same-provider key preservation/propagation; provider changes, Cloud and successful authentication remain untested.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 3,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "replay_setup": {+ "working_directory": "<checkout>",+ "run": "export OH_VERIFY_RUN=<fresh owned run; observed run in live.run>",+ "dummy_profile": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 5,+ "temperature": 0.4+ },+ "neutral_profile": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "profile_body": {+ "llm": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 5,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "launcher_preconditions": {+ "OPENAI_BASE_URL": "http://127.0.0.1:1/v1",+ "OPENAI_API_BASE": "http://127.0.0.1:1/v1"+ },+ "limits": "Two Advanced and one Basic same-provider model-only saves/switches. Basic intentionally drops base_url and therefore needs protected owned fallback routes, restart and doctor. Provider changes, Cloud and successful authentication/inference remain untested. Excluded setup mistake is recorded separately."+ },+ "steps": [+ "control-openhands onboard --skip",+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands api POST /api/profiles/QA_KEY_EDIT --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands browser goto /settings/llm",+ "control-openhands browser click 'testid=profile-row >> has-text=QA_KEY_EDIT >> testid=profile-menu-trigger'",+ "control-openhands browser click 'testid=profile-actions-menu >> testid=profile-edit'",+ "control-openhands browser click 'testid=sdk-section-advanced-toggle'",+ "control-openhands browser attr 'testid=llm-api-key-input' placeholder",+ "control-openhands browser fill 'testid=llm-custom-model-input' openai/gpt-4o-mini",+ "control-openhands browser value 'testid=base-url-input'",+ "control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 1000",+ "control-openhands browser wait-text 'Profile \"QA_KEY_EDIT\" updated' --timeout 60000",+ "control-openhands api GET /api/profiles/QA_KEY_EDIT --pick api_key_set",+ "control-openhands api GET /api/profiles/QA_KEY_EDIT --pick config.model",+ "control-openhands browser goto /conversations/49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands browser click 'testid=chat-input-llm-profile'",+ "control-openhands browser click 'testid=chat-input-llm-profile-option-QA_KEY_EDIT'",+ "control-openhands conversation send 49f5de70-fb5c-42be-a264-03ce103ae56e --prompt 'Reply with exactly pong. Do not run any tools.'",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>')\np=next((run/'private').rglob('49f5de70fb5c42bea26403ce103ae56e/base_state.json'))\nd=json.loads(p.read_text());llm=d['agent']['llm']\nresult={'conversation':d['id'],'model':llm.get('model'),'key_present':bool(llm.get('api_key')),'base_url':llm.get('base_url')}\nq=run/'evidence/F10.edit/key-edit-diagnostic.json';q.parent.mkdir(parents=True,exist_ok=True);q.write_text(json.dumps([result],indent=2)+'\\n')\nprint(result)\nPY",+ "control-openhands conversation pause 49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands api POST /api/profiles/QA_KEY_EDIT_2 --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands browser goto /settings/llm",+ "control-openhands browser click 'testid=profile-row >> has-text=QA_KEY_EDIT_2 >> testid=profile-menu-trigger'",+ "control-openhands browser click 'testid=profile-actions-menu >> testid=profile-edit'",+ "control-openhands browser click 'testid=sdk-section-advanced-toggle'",+ "control-openhands browser attr 'testid=llm-api-key-input' placeholder",+ "control-openhands browser fill 'testid=llm-custom-model-input' openai/gpt-4o-mini",+ "control-openhands browser value 'testid=base-url-input'",+ "control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 1000",+ "control-openhands api GET /api/profiles/QA_KEY_EDIT_2 --pick api_key_set",+ "control-openhands api GET /api/profiles/QA_KEY_EDIT_2 --pick config.model",+ "control-openhands browser network --filter QA_KEY_EDIT_2",+ "control-openhands browser goto /conversations/49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands browser click 'testid=chat-input-llm-profile'",+ "control-openhands browser click 'testid=chat-input-llm-profile-option-QA_KEY_EDIT_2'",+ "control-openhands conversation send 49f5de70-fb5c-42be-a264-03ce103ae56e --prompt 'Reply with exactly pong. Do not run any tools.'",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=next((run/'private').rglob('49f5de70fb5c42bea26403ce103ae56e/base_state.json'));d=json.loads(p.read_text());llm=d['agent']['llm'];result={'conversation':d['id'],'model':llm.get('model'),'key_present':bool(llm.get('api_key')),'base_url':llm.get('base_url'),'profile':'QA_KEY_EDIT_2'};q=run/'evidence/F10.edit/key-edit-diagnostic.json';rows=json.loads(q.read_text());rows.append(result);q.write_text(json.dumps(rows,indent=2)+'\\n');print(result)\nPY",+ "control-openhands conversation pause 49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands api POST /api/profiles/QA_KEY_BASIC --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands browser goto /settings/llm",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\np=Path('backfill-18309-settings-rechecks/<private-fixture>');d=json.loads(p.read_text());d['OPENAI_BASE_URL']='http://127.0.0.1:1/v1';d['OPENAI_API_BASE']='http://127.0.0.1:1/v1';p.write_text(json.dumps(d));print({'owned_default_OPENAI_BASE_URL_is_loopback':True,'owned_default_OPENAI_API_BASE_is_loopback':True,'AWS_EC2_METADATA_DISABLED':d.get('AWS_EC2_METADATA_DISABLED')=='true'})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands browser reload",+ "control-openhands browser click 'testid=profile-row >> has-text=QA_KEY_BASIC >> testid=profile-menu-trigger'",+ "control-openhands browser click 'testid=profile-actions-menu >> testid=profile-edit'",+ "control-openhands browser click 'testid=sdk-section-basic-toggle'",+ "control-openhands browser attr 'testid=llm-api-key-input' placeholder",+ "control-openhands browser choose 'testid=llm-model-input' gpt-5.6",+ "control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 1000",+ "control-openhands browser wait-text 'Profile \"QA_KEY_BASIC\" updated' --timeout 60000",+ "control-openhands api GET /api/profiles/QA_KEY_BASIC --pick api_key_set",+ "control-openhands api GET /api/profiles/QA_KEY_BASIC --pick config.model",+ "control-openhands api GET /api/profiles/QA_KEY_BASIC --pick config.base_url",+ "control-openhands browser goto /conversations/49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands browser click 'testid=chat-input-llm-profile'",+ "control-openhands browser click 'testid=chat-input-llm-profile-option-QA_KEY_BASIC'",+ "control-openhands conversation send 49f5de70-fb5c-42be-a264-03ce103ae56e --prompt 'Reply with exactly pong. Do not run any tools.'",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=next((run/'private').rglob('49f5de70fb5c42bea26403ce103ae56e/base_state.json'));d=json.loads(p.read_text());llm=d['agent']['llm'];r={'conversation':d['id'],'profile':'QA_KEY_BASIC','view':'basic','model':llm.get('model'),'key_present':bool(llm.get('api_key')),'base_url':llm.get('base_url'),'owned_fallback_endpoint':'http://127.0.0.1:1/v1'};q=run/'evidence/F10.edit/key-edit-diagnostic.json';rows=json.loads(q.read_text());rows.append(r);q.write_text(json.dumps(rows,indent=2)+'\\n');print(r)\nPY",+ "control-openhands conversation pause 49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands browser screenshot --feature F10.edit --name 17806-basic-key-preserved"+ ],+ "evidence": [+ "backfill-18309-settings-rechecks/evidence/F10.edit/key-edit-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F10.edit/17806-dummy-key-preserved.png",+ "backfill-18309-settings-rechecks/evidence/F10.edit/17806-basic-key-preserved.png"+ ],+ "observation": "All three actual saved/switched profiles kept the dummy key; Basic runtime base_url was null and its backend default endpoint was protected loopback.",+ "setup_mistakes": [+ "One premature Basic Save removed base_url and reached OpenAI with only the dummy key; validation rejected AuthenticationError, no profile save or inference succeeded. This setup mistake is excluded from product verdict/proof. The corrected Basic run was performed only after owned default routes, restart and doctor."+ ],+ "initial_profile_requests": {+ "file": "backfill-18309-settings-rechecks/<private-fixture>",+ "names": [+ "QA_KEY_EDIT",+ "QA_KEY_EDIT_2",+ "QA_KEY_BASIC"+ ],+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 5,+ "temperature": 0.4+ },+ "include_secrets": true+ }+ },+ "changes": [+ {+ "name": "QA_KEY_EDIT",+ "view": "Advanced",+ "from": "openai/gpt-4o",+ "to": "openai/gpt-4o-mini"+ },+ {+ "name": "QA_KEY_EDIT_2",+ "view": "Advanced",+ "from": "openai/gpt-4o",+ "to": "openai/gpt-4o-mini"+ },+ {+ "name": "QA_KEY_BASIC",+ "view": "Basic",+ "from": "openai/gpt-4o",+ "to": "openai/gpt-5.6"+ }+ ],+ "replay_notes": [+ "Use freshly returned conversation IDs and fixture paths. Create QA_KEY_EDIT_2 only after the first edit to keep the historical row selector unambiguous.",+ "Read the second save result after its bounded 30s client deadline; captured browser network had POST 201 at 30.007s. The missed transient second toast is not proof.",+ "The excluded first Basic setup attempt and its rejected provider validation appear only in setup_mistakes, not in successful proof steps."+ ],+ "verified_owned_server_environment": {+ "OPENAI_BASE_URL": "http://127.0.0.1:1/v1",+ "OPENAI_API_BASE": "http://127.0.0.1:1/v1",+ "AWS_EC2_METADATA_DISABLED": "true",+ "confirmed_as_booleans_only": true+ },+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ },+ "run_id": "backfill-18309-settings-rechecks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17806#issuecomment-6122466133"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17803,+ "title": "[Bug]: Editing provider-connection base_url does not take effect on next conversation (stale snapshot in settings.json)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17803",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F11",+ "feature_ids": [+ "F11.edit",+ "F11.agent-uses-connection"+ ],+ "basis": "live-reproduced",+ "reason": "Two real connection Base URL edits persisted /v2 then /v3, but two fresh Canvas-created conversations still materialized the activated /v1 URL. The stored connection, settings snapshot and run-owned runtime state were read separately. This proves stale URL resolution before model success; all URLs were refused loopback.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "replay_setup": {+ "working_directory": "<checkout>",+ "run": "export OH_VERIFY_RUN=<fresh owned run; observed run in live.run>",+ "dummy_profile": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ }+ },+ "steps": [+ "control-openhands api POST /api/llm/provider-connections --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands api POST /api/profiles/QA_URL_LINKED --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands api POST /api/profiles/QA_URL_LINKED/activate --write",+ "control-openhands browser goto /settings/llm",+ "control-openhands browser click 'testid=provider-connection-row >> has-text=QA_URL_PATCH >> testid=provider-connection-menu-trigger'",+ "control-openhands browser testids --filter provider-connection",+ "control-openhands browser click 'testid=provider-connection-edit'",+ "control-openhands browser fill 'testid=provider-connection-base-url-input' http://127.0.0.1:1/v2",+ "control-openhands browser click 'testid=provider-connection-submit'",+ "control-openhands api GET /api/settings --pick agent_settings.llm.base_url",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');cid='4d60ac0e-9833-4475-a187-bb80dcc4de82';p=next((run/'private').rglob(cid.replace('-','')+'/base_state.json'));d=json.loads(p.read_text());llm=d['agent']['llm'];r={'conversation':cid,'connection_url_after_ui_save':'http://127.0.0.1:1/v2','runtime_url':llm.get('base_url'),'model':llm.get('model'),'key_present':bool(llm.get('api_key'))};q=run/'evidence/F11.agent-uses-connection/url-edit-diagnostic.json';q.parent.mkdir(parents=True,exist_ok=True);q.write_text(json.dumps([r],indent=2)+'\\n');print(r)\nPY",+ "control-openhands api GET /api/llm/provider-connections",+ "control-openhands doctor",+ "control-openhands browser goto /settings/llm",+ "control-openhands browser click 'testid=provider-connection-row >> has-text=QA_URL_PATCH >> testid=provider-connection-menu-trigger'",+ "control-openhands browser click 'testid=provider-connection-edit'",+ "control-openhands browser value 'testid=provider-connection-base-url-input'",+ "control-openhands browser fill 'testid=provider-connection-base-url-input' http://127.0.0.1:1/v3",+ "control-openhands browser click 'testid=provider-connection-submit'",+ "control-openhands api GET /api/llm/provider-connections",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');cid='b299a4b6-5622-49a2-a962-3b87ef189190';p=next((run/'private').rglob(cid.replace('-','')+'/base_state.json'));d=json.loads(p.read_text());llm=d['agent']['llm'];r={'conversation':cid,'connection_url_after_ui_save':'http://127.0.0.1:1/v3','runtime_url':llm.get('base_url'),'model':llm.get('model'),'key_present':bool(llm.get('api_key'))};q=run/'evidence/F11.agent-uses-connection/url-edit-diagnostic.json';rows=json.loads(q.read_text());rows.append(r);q.write_text(json.dumps(rows,indent=2)+'\\n');print(r)\nPY",+ "control-openhands conversation status b299a4b6-5622-49a2-a962-3b87ef189190",+ "control-openhands browser screenshot --feature F11.agent-uses-connection --name 17803-stale-loopback-url",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write"+ ],+ "evidence": [+ "backfill-18309-settings-rechecks/evidence/F11.agent-uses-connection/url-edit-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F11.agent-uses-connection/17803-stale-loopback-url.png"+ ],+ "observation": "Stored connection URLs changed; each newly initialized conversation used the old activated URL.",+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ },+ "run_id": "backfill-18309-settings-rechecks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17801,+ "title": "[Bug]: Adding or Editing an LLM does not always take you to the Basic tab",+ "url": "https://github.com/OpenHands/OpenHands/issues/17801",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.create-prefill",+ "F10.edit"+ ],+ "basis": "live-reproduced",+ "reason": "Add and Edit selected All twice for a profile with a custom base URL and temperature 0.4, rather than the requested Basic starting tab.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server 1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Add and Edit selected All twice for a profile with a custom base URL and temperature 0.4, rather than the requested Basic starting tab.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F10.create-prefill/17801-add-all.png",+ "backfill-18309-settings-worker/evidence/F10.edit/17801-edit-all.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17788,+ "title": "[Bug] Cloud attachment upload rejects valid runtime data when cached fields are empty strings",+ "url": "https://github.com/OpenHands/OpenHands/issues/17788",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.attach-files"+ ],+ "basis": "known-issue",+ "reason": "A Cloud sandbox whose cached runtime URL/key are blank while the fresh runtime lookup is already ready.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17785,+ "title": "[Bug]: Windows Desktop v1.24.0 - Saving DeepSeek-v4-flash LLM config throws AttributeError: 'PromptTokensDetailsWrapper' object has no attribute 'cache_creation_tokens'",+ "url": "https://github.com/OpenHands/OpenHands/issues/17785",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.create-validation"+ ],+ "basis": "known-issue",+ "reason": "Saving DeepSeek-v4-flash on Windows Desktop can fail with missing cache_creation_tokens instead of saving. The reported path needs a Windows Desktop 1.24.0 environment with its SDK dependency resolution and an approved DeepSeek key; this macOS source run cannot establish that result.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17783,+ "title": "[Bug]: Canvas App install form throws on malformed percent escapes in tree URLs",+ "url": "https://github.com/OpenHands/OpenHands/issues/17783",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F20",+ "feature_ids": [+ "F20.git-tree-url",+ "F20.install-error"+ ],+ "basis": "live-reproduced",+ "reason": "Executed both malformed ref and path tree URLs in the live install modal; each raised URI malformed, no install mutation or toast, modal remains. Reviewed local demo control installs and enables.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Malformed tree-URL escapes (`F20.git-tree-url`, `F20.install-error`).** On `/apps`, open `canvas-extensions-add-button`, fill `add-canvas-extension-source-input` with `https://github.com/o/r/tree/main%/apps/demo`, and click `add-canvas-extension-submit`. Inspect `browser errors --app-only`, `browser toasts --history`, and the modal count. Repeat with `https://gitlab.com/group/repo/-/tree/main/apps/%E0%A4%A`. Expected: submission produces a usable install/validation outcome. Known failure, reproduced 2026-10-12 on `fa1f724` with Agent Server 1.54.0: each submit raises `URI malformed`, emits no toast, skips the install request, and leaves the modal open ([#17783](https://github.com/OpenHands/OpenHands/issues/17783)). A reviewed local demo source still installs successfully as the control; no malformed source is executed."+ ],+ "artifacts": [+ "evidence/F20.git-tree-url/malformed-ref-17783.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17774,+ "title": "[Bug]: Planner tab fails with \"Conversation file path must stay inside the workspace\" on Windows local backend",+ "url": "https://github.com/OpenHands/OpenHands/issues/17774",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.planner-empty",+ "F27.planner-plan"+ ],+ "basis": "known-issue",+ "reason": "Native Windows local Agent Server and drive-letter workspace fixtures are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17773,+ "title": "[Bug]: Files tab always shows \"No files in workspace\" on Windows local backend",+ "url": "https://github.com/OpenHands/OpenHands/issues/17773",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-tree"+ ],+ "basis": "known-issue",+ "reason": "A Windows local Agent Server using cmd.exe; this run is macOS.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17763,+ "title": "[Bug]: Self-hosted LOCAL_BACKEND_API_KEY is not exported to OH_SESSION_API_KEYS_0 in entrypoint.sh, causing agent-server auth bypass",+ "url": "https://github.com/OpenHands/OpenHands/issues/17763",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.docker-image"+ ],+ "basis": "known-issue",+ "reason": "An owned Docker container/daemon is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17710,+ "title": "[Bug]: Secret edit/deletion leads to duplicate secret being created and none being deleted",+ "url": "https://github.com/OpenHands/OpenHands/issues/17710",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F14",+ "feature_ids": [+ "F14.delete",+ "F14.edit"+ ],+ "basis": "known-issue",+ "reason": "On the Cloud backend, deleting a secret reportedly adds a suffixed duplicate and leaves the original. Reproduction needs a disposable writable OpenHands Cloud account at the affected service; Local secret CRUD exercises a different transport.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F14-secrets.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17706,+ "title": "[Bug]: Canvas conversations on openhands/claude-* fail with 400 on prompt_cache_key (uvx resolves litellm>=1.99)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17706",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.default-used"+ ],+ "basis": "known-issue",+ "reason": "A freshly resolved npm-launcher environment can reject openhands/claude-* conversations because LiteLLM sends prompt_cache_key. Reproduction needs an approved OpenHands proxy key with an Anthropic model and the affected fresh transitive dependency resolution; a keyless initialization cannot prove provider acceptance.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17704,+ "title": "[Bug]: A non-ASCII file name in a script bundle shows garbled (café.py as café.py) and loses its Entrypoint badge",+ "url": "https://github.com/OpenHands/OpenHands/issues/17704",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.script-section"+ ],+ "basis": "live-reproduced",+ "reason": "Owned UTF-8 USTAR bundle café.py/main.py, inactive automation python3 café.py: actual Script header renders café.py and no Entrypoint badge; ASCII control untouched.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **UTF-8 script member (`F23.script-section`).** Arrange an inactive annual-cron script automation through the real upload/create APIs with a two-file UTF-8 USTAR bundle (`tarfile.USTAR_FORMAT`, names `café.py` and `main.py`) and entrypoint `python3 café.py`; do not use a PAX-only member name. Open `/automations/<id>` and inspect `automation-script-entrypoint`, both `automation-script-file` headings, and the screenshot. Expected: `café.py` keeps its name, receives the **Entrypoint** badge and sorts first. Known failure, reproduced 2026-10-12 on `fa1f724`: the command is correct, but the file heading is `café.py` without the badge; the ASCII control reads normally ([#17704](https://github.com/OpenHands/OpenHands/issues/17704)). No run or model is needed."+ ],+ "artifacts": [+ "evidence/F23.script-section/unicode-17704.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17657,+ "title": "fix(ci): repair release ready workflow startup failures",+ "url": "https://github.com/OpenHands/OpenHands/issues/17657",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Release-ready workflow permission/startup failure; repository CI tooling."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17623,+ "title": "[Feature]: Helm chart should not automount the ServiceAccount token in the agent pod by default",+ "url": "https://github.com/OpenHands/OpenHands/issues/17623",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.helm-chart"+ ],+ "basis": "known-issue",+ "reason": "Helm and a disposable Kubernetes cluster are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17622,+ "title": "[Bug]: Node.js version in README/SELF_HOSTING docs (22.12.x) contradicts engines.node >=24 and CI (24.15)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17622",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Node requirement documentation drift; documentation only."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17621,+ "title": "[Bug]: Compact context always fails after exactly 60s — client-side timeout aborts a condensation the server then completes",+ "url": "https://github.com/OpenHands/OpenHands/issues/17621",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "basis": "known-issue",+ "reason": "A real summarizing model and conversation whose condensation takes longer than 60 seconds.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17620,+ "title": "[Bug]: Fresh-clone npm run typecheck fails with TS2307 for #/i18n/declaration (make-i18n not chained)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17620",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Fresh-clone typecheck generation order; contributor build tooling."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17619,+ "title": "[Bug]: WebSocket reconnect resubscribes from a stale `after_timestamp`, replaying the whole session",+ "url": "https://github.com/OpenHands/OpenHands/issues/17619",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "family": "F06",+ "feature_ids": [+ "F06.reconnect"+ ],+ "basis": "known-issue",+ "reason": "The initial session socket used after_seq=-1. Three genuine run:false user messages yielded events/count=7; the same open page reconnected after an owned backend restart with after_seq=6. A recovery message appeared once. A recorded repeat advanced -1 to2 with count3. No frame bodies were captured, so this does not quantify server replayed payloads or cost.",+ "steps": [+ "Recorded actions and replay guidance: export OH_VERIFY_RUN to the full directory returned by a new isolated launch; replace conversation IDs and fixture paths with those returned by that run. The recorded run and measurements below are unchanged. JSON bodies shown after --data are shell-quoted; fixture JSON must be saved before an @file command.",+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser goto /conversations/<id>",+ "control-openhands --run \"$OH_VERIFY_RUN\" api POST /api/conversations/<id>/events --write --data '{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"QA_CURSOR_17619_<1..3>\"}],\"run\":false}' (one request per unique marker)",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser wait-text QA_CURSOR_17619_3; control-openhands --run \"$OH_VERIFY_RUN\" api GET /api/conversations/<id>/events/count (observed7, latest sequence6)",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser events --kinds websocket-created,websocket-open,websocket-close,websocket-error --last 100 (initial after_seq=-1)",+ "control-openhands --run \"$OH_VERIFY_RUN\" restart; control-openhands --run \"$OH_VERIFY_RUN\" doctor. Keep Chromium and the open page; do not reload. Confirm actual close and a new communicating connection with after_seq=6.",+ "control-openhands --run \"$OH_VERIFY_RUN\" api POST /api/conversations/<id>/events --write --data '{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"QA_CURSOR_17619_RECOVERY\"}],\"run\":false}'; control-openhands --run \"$OH_VERIFY_RUN\" browser wait-text QA_CURSOR_17619_RECOVERY; control-openhands --run \"$OH_VERIFY_RUN\" browser count text=\"QA_CURSOR_17619_RECOVERY\" (observed1)",+ "A separate recorded repeat used browser record start --feature F06.reconnect --name cursor-recovery-17619, one before marker, events/count=3, restart without page reload, one recovery marker, then browser record stop --keep-frames."+ ],+ "actual": "The initial session socket used after_seq=-1. Three genuine run:false user messages yielded events/count=7; the same open page reconnected after an owned backend restart with after_seq=6. A recovery message appeared once. A recorded repeat advanced -1 to2 with count3. No frame bodies were captured, so this does not quantify server replayed payloads or cost.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/socket-cursor-17619.json",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/socket-cursor-video-17619.json",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/cursor-recovery-17619.mp4",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/recovered-video-17619.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "run_id": "backfill-18309-surface-rechecks",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17619#issuecomment-6122465227"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17577,+ "title": "[Bug]: Basic LLM form cannot use a model the provider catalog does not list (ready-for-dev companion to #17233)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17577",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.model-picker"+ ],+ "basis": "live-reproduced",+ "reason": "Basic OpenRouter discarded the off-catalog model ID on blur in two trials; the model field became empty.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server 1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Basic OpenRouter discarded the off-catalog model ID on blur in two trials; the model field became empty.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F10.model-picker/17577-off-catalog.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17567,+ "title": "[Bug]: Switching from expanded Git diff to Planner raises a Monaco disposed-model exception",+ "url": "https://github.com/OpenHands/OpenHands/issues/17567",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.diff-view-modes"+ ],+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17566,+ "title": "[Bug]: Terminal tab loses command history after page reload while chat retains it",+ "url": "https://github.com/OpenHands/OpenHands/issues/17566",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.terminal-history-reload"+ ],+ "basis": "existing-note",+ "reason": "Existing terminal history failure reproduced 2026-10-06 is preserved; no available model credential for a new terminal command.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17564,+ "title": "[Bug]: Mobile automation dashboard actions squeeze the heading to about 52px",+ "url": "https://github.com/OpenHands/OpenHands/issues/17564",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "not-reproduced",+ "family": "F21",+ "feature_ids": [+ "F21.phone"+ ],+ "reason": "With 25 inert automation fixtures, the dashboard heading measures 347 px at 390 px and 277 px at 320 px, at 28 px height. Actions stack below it and no horizontal page overflow appears. This exact current dashboard layout does not reproduce the reported 52 px squeeze.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Launch a fresh private run at fa1f724, doctor, browser start, onboard --skip.",+ "Arrange 25 inert annual-cron automations through /api/automation/v1 with state INACTIVE except two bounded script fixtures.",+ "browser goto /automations; browser viewport phone; browser bbox 'role=heading[name=\"Dashboard\"]'; browser screenshot --feature F21.phone --name header-390-17564.",+ "Repeat browser viewport 320x700 and bbox/screenshot; read document.documentElement.scrollWidth and innerWidth through read-only browser eval.",+ "Observed heading widths 347 px at 390 px and 277 px at 320 px, 28 px height; actions stack below heading and no horizontal overflow."+ ],+ "artifacts": [+ "evidence/F21.phone/header-390-17564.png",+ "evidence/F21.phone/header-320-17564.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17564#issuecomment-6121791800"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17563,+ "title": "[Bug]: Mobile automation edit dialog clips its header and footer",+ "url": "https://github.com/OpenHands/OpenHands/issues/17563",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.phone",+ "F23.edit-close"+ ],+ "basis": "existing-note",+ "reason": "Existing mobile clipping and Escape notes preserved.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17559,+ "title": "[Bug]: History pagination can show both an action and its observation (ready-for-dev companion to #16947)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17559",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.load-older-history"+ ],+ "basis": "known-issue",+ "reason": "A real persisted history page boundary splitting an action/observation or ACP started/completed pair; dummy-key errors do not produce these tool pairs.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17546,+ "title": "[Bug]: detect ports automatically instead of always using 3001",+ "url": "https://github.com/OpenHands/OpenHands/issues/17546",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.cli-port-in-use"+ ],+ "basis": "live-reproduced",+ "reason": "Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. It provides no automatic frontend fallback, and the ingress override does not change that internal port.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Occupied internal frontend port (`F26.cli-port-in-use`).** Reserve a loopback frontend port in an owned listener and run the genuine launcher with separate free ingress/backend/automation ports. The reported default-port branch can also be observed read-only when port 3001 is already occupied; do not stop its owner. Expected: an occupied internal frontend port receives a free fallback while explicit service-port conflicts keep their existing semantics. Known failure, reproduced 2026-10-12: `agent-canvas --port 18940` exits 1 with `frontend: port 3001`, despite the ingress override ([#17546](https://github.com/OpenHands/OpenHands/issues/17546), [#16226](https://github.com/OpenHands/OpenHands/issues/16226)). An explicit `OH_CANVAS_SAFE_VITE_PORT` is a current workaround; it does not provide automatic fallback."+ ],+ "artifacts": [+ "evidence/F26.cli-port-in-use/frontend-default-17546-16226.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17520,+ "title": "[Bug]: Deleting the title LLM profile leaves a stale row when settings cleanup fails",+ "url": "https://github.com/OpenHands/OpenHands/issues/17520",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.delete"+ ],+ "basis": "live-reproduced",+ "reason": "Two deletes returned 200 while title-cleanup PATCH returned 500; the UI kept the deleted row although its API returned 404. Temporary own-file immutable flags were restored immediately and doctor passed.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Two deletes returned 200 while title-cleanup PATCH returned 500; the UI kept the deleted row although its API returned 404. Temporary own-file immutable flags were restored immediately and doctor passed.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F10.delete/17520-immutable-cleanup-error.png",+ "backfill-18309-settings-worker/evidence/F10.delete/17520-immutable-repeat.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17508,+ "title": "Agent Canvas update banner keeps showing 1.19.0 after npm install -g @openhands/agent-canvas@latest (installed package is 1.20.0)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17508",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F09",+ "feature_ids": [+ "F09.update-available"+ ],+ "basis": "known-issue",+ "reason": "Isolated old/new package installs with the old process still serving, or the original stale bundle/cache capture after update.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F09-settings-shell.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17497,+ "title": "[Bug]: Corrupted conversation data breaks the entire conversation history list instead of handling individual broken conversations",+ "url": "https://github.com/OpenHands/OpenHands/issues/17497",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "basis": "known-issue",+ "reason": "The original corrupted persisted conversation or a deterministic corruption fixture after model-switch failure and fork.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17494,+ "title": "[Bug]: Transcript export leaks raw i18n template for ACP tool calls without a title",+ "url": "https://github.com/OpenHands/OpenHands/issues/17494",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.export-transcript"+ ],+ "basis": "known-issue",+ "reason": "An authenticated ACP session that emits a titleless tool call, then Markdown and HTML export.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17492,+ "title": "Failed browser, glob, grep, and file editor tool calls render as successful in chat",+ "url": "https://github.com/OpenHands/OpenHands/issues/17492",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A working model emitting failed browser, glob, grep, planning-editor and file-editor observations with is_error=true.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17490,+ "title": "Terminal commands that fail with an ordinary non-zero exit render as successful in chat",+ "url": "https://github.com/OpenHands/OpenHands/issues/17490",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A working model producing a completed terminal command with nonzero exit_code and is_error=false.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17485,+ "title": "Git info probe on the shared bash-events socket can resolve with another client's command output",+ "url": "https://github.com/OpenHands/OpenHands/issues/17485",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "The reported persistent shared bash-events correlation consumer was removed. Current useBashCommandRunner uses conversation-scoped BashClient.executeCommand."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17476,+ "title": "Planning socket stats overwrite the main conversation's context meter",+ "url": "https://github.com/OpenHands/OpenHands/issues/17476",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "basis": "known-issue",+ "reason": "A real planning sub-conversation and main conversation with distinct nonzero usage stats.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17471,+ "title": "[Bug]: Chat attachment picker silently ignores re-picking the same file after removing it",+ "url": "https://github.com/OpenHands/OpenHands/issues/17471",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.attach-files"+ ],+ "basis": "live-reproduced",+ "reason": "Same-file re-pick after removing its chip leaves attachment count zero.",+ "steps": [+ "fixture file --name qa-surface.txt --content \"feature-map backfill attachment\"",+ "browser goto /",+ "browser upload testid=upload-image-input <run>/evidence/_fixtures/qa-surface.txt",+ "browser click role=button[name=\"Remove qa-surface.txt\"]",+ "browser upload testid=upload-image-input <same file>",+ "browser count role=button[name=\"Remove qa-surface.txt\"] => 0"+ ],+ "evidence": [+ "evidence/F05.attach-files/issue-17471-repick.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17470,+ "title": "Agent server crashes with SIGILL on Apple Silicon under Colima (OpenSSL ARM capability misdetection)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17470",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.docker-image"+ ],+ "basis": "known-issue",+ "reason": "Affected Apple Silicon Colima ARM container environment and Docker daemon are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17436,+ "title": "Prompt-preset automations created on agent-canvas 1.17.0 fail every run with 500 on /api/conversations/{id}/events (preset template finish-tool wiring vs. pinned agent-server 1.46.0)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17436",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F22",+ "feature_ids": [+ "F22.setup-review-confirm"+ ],+ "basis": "known-issue",+ "reason": "A real LLM key and the reported legacy release/template pair are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17429,+ "title": "[Bug]: 500 error with /btw",+ "url": "https://github.com/OpenHands/OpenHands/issues/17429",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.slash-btw"+ ],+ "basis": "known-issue",+ "reason": "The exact pending tool/streaming state that triggers the generator-pickle error, plus a working model for the side question.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17426,+ "title": "[Bug]: Nit: agent profile says \"set as active\" instead of \"set as default\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/17426",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.set-active"+ ],+ "basis": "live-reproduced",+ "reason": "The active agent profile badge says Default while its menu says Set as active, on two openings.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server 1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "The active agent profile badge says Default while its menu says Set as active, on two openings.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F13.set-default/17426-wording.png",+ "backfill-18309-settings-worker/evidence/F13.set-default/17426-wording.aria.txt"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17391,+ "title": "[Bug]: ChatGPT Business Subscriptions",+ "url": "https://github.com/OpenHands/OpenHands/issues/17391",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.subscription"+ ],+ "basis": "known-issue",+ "reason": "ChatGPT Business subscription login can be refused or use a fallback model with a 404. Completing the entitlement and selected-model checks needs an approved ChatGPT Business workspace login and the reported Docker subscription runtime.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17390,+ "title": "HELP: CLOUD",+ "url": "https://github.com/OpenHands/OpenHands/issues/17390",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.cloud-sandbox-states"+ ],+ "basis": "known-issue",+ "reason": "Affected authenticated Cloud account/session and token-provisioning reproduction are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17386,+ "title": "[Bug]: Workspace grouping disappears across browsers or after clearing site data",+ "url": "https://github.com/OpenHands/OpenHands/issues/17386",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.grouped-folders"+ ],+ "basis": "live-reproduced",+ "reason": "Fresh browser puts existing workspace-bound conversations into No workspace.",+ "steps": [+ "Create and register a workspace, then create empty conversations from its sidebar launch-workspace entry.",+ "View > layout-preset-by-workspace; capture groups.",+ "browser reset; onboard --skip; View > layout-preset-by-workspace.",+ "browser snapshot conversation-panel => prior conversations under No workspace; registered workspace remains.",+ "conversation status <id> readback retains workspace.working_dir matching the registered folder."+ ],+ "evidence": [+ "evidence/F04.grouped-folders/issue-17386-after-reset.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17384,+ "title": "[Bug]: File-based subagents use encrypted provider API key without decrypting it",+ "url": "https://github.com/OpenHands/OpenHands/issues/17384",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.launch-child"+ ],+ "basis": "known-issue",+ "reason": "A real encrypted provider connection/profile, reviewed file-based worker fixture and authorized model key are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17383,+ "title": "[Bug]: Browser tooling fails due to incompatible MCP dependency version",+ "url": "https://github.com/OpenHands/OpenHands/issues/17383",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.browser-screenshot"+ ],+ "basis": "known-issue",+ "reason": "Actual pinned runtime browser-use 0.11.13/mcp 1.30.0/fastmcp 3.4.8 BrowserToolExecutor constructs. The report’s supported Canvas/global browser operation still needs a genuine browser-enabled agent launch and authorized LLM to issue a browser_* operation; a constructor probe alone does not establish that user flow.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Actual pinned runtime browser-use 0.11.13/mcp 1.30.0/fastmcp 3.4.8 BrowserToolExecutor constructs. The report’s supported Canvas/global browser operation still needs a genuine browser-enabled agent launch and authorized LLM to issue a browser_* operation; a constructor probe alone does not establish that user flow."+ ],+ "artifacts": [+ "evidence/F27.browser-screenshot/dependency-17383-5152.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17371,+ "title": "[Bug]: rotated session key isn't reconciled into openhands-backends → silent permanent 401 loop",+ "url": "https://github.com/OpenHands/OpenHands/issues/17371",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.add-agent-server",+ "F25.recovery-gate"+ ],+ "basis": "live-reproduced",+ "reason": "Added QA_Rotation_Legacy for the identical origin through Add Backend using the owned session-key file. After restart --rotate-key and reload, read-only comparison shows default-local matches injected key but the manually registered active entry does not. UI shows Disconnected and Unauthorized; core unreconciled credential persists, although there is visible status feedback in this version.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Manually registered same-origin key rotation (`F25.add-agent-server`, `F25.recovery-gate`).** Through **Add Backend → Agent-server**, add an owned entry named `QA_Rotation_Legacy` with the run's exact origin and fill its API Key using `--value-file <run>/private/session-key`; keep it active. Run `restart --rotate-key`, `doctor`, and `browser reload`. Compare only each saved entry's `apiKey === window.__AGENT_CANVAS_SESSION_API_KEY__` boolean. Expected: same-origin entries accept the newly injected session key and the active backend recovers. Known failure, reproduced 2026-10-12: `default-local` matches the injected key, the manually registered entry does not, and the UI shows **Disconnected**/**Unauthorized** ([#17371](https://github.com/OpenHands/OpenHands/issues/17371)). This version has visible status feedback, so do not repeat the older claim of a completely silent loop. Choose **Connected Local** to recover, then remove the owned duplicate."+ ],+ "artifacts": [+ "evidence/F25.recovery-gate/rotated-manual-backend-17371.png",+ "evidence/F25.recovery-gate/rotated-key-booleans-17371.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17271,+ "title": "[Bug]: New chat retains context/state from the previous conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/17271",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "basis": "known-issue",+ "reason": "A completed conversation with nonzero token usage before New Chat; a dummy-key run supplies no genuine context usage.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17263,+ "title": "[Bug]: Context meter stays pinned after a successful compaction once the condenser has its own usage entry",+ "url": "https://github.com/OpenHands/OpenHands/issues/17263",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "basis": "known-issue",+ "reason": "A real successful compaction whose condenser usage remains larger than the default agent usage.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17261,+ "title": "[Bug]: Cannot add LLM profile with advanced settings",+ "url": "https://github.com/OpenHands/OpenHands/issues/17261",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.custom-model",+ "F10.connection-link"+ ],+ "basis": "live-reproduced",+ "reason": "A real Advanced UI Save of the exact unqualified qwen model linked to an OpenAI connection succeeded; both subsequent actual Canvas conversations emitted LLM Provider NOT provided model=qwen before HTTP. The reported Windows installation, unlinked preflight and successful local inference remain untested.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "steps": [+ "control-openhands browser goto /settings/llm",+ "control-openhands browser click 'testid=add-llm-profile'",+ "control-openhands browser click 'testid=sdk-section-advanced-toggle'",+ "control-openhands browser fill 'testid=profile-name-input' QA_QWEN",+ "control-openhands browser fill 'testid=llm-custom-model-input' qwen",+ "control-openhands browser choose 'testid=llm-provider-connection-input' QA_URL_PATCH",+ "control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 1000",+ "control-openhands api GET /api/profiles/QA_QWEN --pick config.model",+ "control-openhands api POST /api/profiles/QA_QWEN/activate --write",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events cbad8891-d64c-4998-999a-b8f7b7f50f60 --kinds ConversationErrorEvent --last 2",+ "control-openhands browser screenshot --feature F10.custom-model --name 17261-qwen-dispatch",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events 0f4774cd-9692-4389-8050-8c34195952d8 --kinds ConversationErrorEvent --last 2",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write"+ ],+ "evidence": [+ "backfill-18309-settings-rechecks/evidence/F10.custom-model/17261-qwen-dispatch.png",+ "backfill-18309-settings-rechecks/evidence/F10.custom-model/qwen-dispatch-diagnostic.json"+ ],+ "observation": "Both actual starts failed at provider selection for saved qwen despite a configured OpenAI connection.",+ "fixture_specs": {+ "model": "qwen",+ "connection": {+ "display_name": "QA_URL_PATCH",+ "provider": "openai",+ "base_url": "http://127.0.0.1:1/v3"+ }+ },+ "uncovered_branches": [+ "exact Windows installation",+ "successful local inference",+ "unlinked inline preflight branch"+ ],+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ },+ "run_id": "backfill-18309-settings-rechecks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17257,+ "title": "useStartTasks returns a hardcoded empty array — in-progress start-task cards have not rendered since #278",+ "url": "https://github.com/OpenHands/OpenHands/issues/17257",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.start-task-cards"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Cloud backend producing an in-progress sandbox/repository start task.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17239,+ "title": "[Bug]: model picker and LLM-profile picker clipped at top of viewport",+ "url": "https://github.com/OpenHands/OpenHands/issues/17239",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.llm-profile-picker"+ ],+ "basis": "known-issue",+ "reason": "Local Home/conversation profile-picker checks passed, but the reported Cloud and real ACP model-picker branches need an approved Cloud or ACP session with a multi-row model catalogue.",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server 1.54.0",+ "viewport": [+ "1512x720",+ "1440x1000",+ "390x844",+ "320x700"+ ],+ "passes": 2,+ "steps": [+ "Arrange exactly eight QA profiles through real API with dummy key and loopback127.0.0.1:1/v1; activate QA_LOOPBACK.",+ "Home1512x720 open chat-input-llm-profile: popover bbox{x574,y6,w200,h421.6}, scrollHeight521/clientHeight420. First row inside.",+ "Scroll chat-input-llm-profile-popover by520: scrollTop101, last row y334/h54 fully inside. Click QA_PROFILE_07; pill says QA_PROFILE_07. Reopen scroll by-520 and select QA_LOOPBACK. Repeat at1512x720.",+ "Desktop menu x538/y6/w200/h500; last rowy435/h54 inside.",+ "Phone menux68/y87.9/w200/h506.4; scrollTop17; lastrowy499.9/h54 inside and clickchangespill.",+ "Narrow Home remains inline; popoverx68/y158/w200/h420, scrollTop103; lastrowy484/h54inside and selectable.",+ "Restore desktop and QA_LOOPBACK.",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60 (dummy loopback only; initialization ends in connection refusal, genuine conversation route remains usable).",+ "control-openhands browser viewport 1512x720; browser click testid=chat-input-llm-profile; browser bbox testid=chat-input-llm-profile-popover: x574,y196,w200,h432, insideViewporttrue.",+ "control-openhands browser scroll testid=chat-input-llm-profile-popover --by 520; browser bbox testid=chat-input-llm-profile-option-QA_PROFILE_07: y534,h54, insideViewporttrue.",+ "control-openhands browser click testid=chat-input-llm-profile-option-QA_PROFILE_07; browser text testid=chat-input-llm-profile returnsQA_PROFILE_07; reopen and scroll--by -520; selectQA_LOOPBACK.",+ "control-openhands browser viewport phone; openpill; menux68,y245.6,w200,h506.4insideViewport; scroll--by 520 to17; lastrowy657.6,h54inside; selectlast and thenfirst.",+ "control-openhands browser viewport desktop (restored)."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F05.model-picker/17239-eight-1512x720.png",+ "backfill-18309-settings-worker/evidence/F05.model-picker/17239-eight-desktop.png",+ "backfill-18309-settings-worker/evidence/F05.model-picker/17239-eight-phone-scrolled.png",+ "backfill-18309-settings-worker/evidence/F05.llm-profile-picker/17239-eight-narrow-scrolled.png",+ "backfill-18309-settings-worker/evidence/F05.llm-profile-picker/17239-conversation-1512x720.png",+ "backfill-18309-settings-worker/evidence/F05.llm-profile-picker/17239-conversation-phone.png"+ ],+ "conversation_id": "fda5a4cd-9c39-4091-8b9a-5512f4c48642",+ "result": "pass",+ "actual": "Eight Local profile rows remained reachable through scroll and first/last selections; untested branches remain blocked.",+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17233,+ "title": "[Bug]: Openrouter custom model support",+ "url": "https://github.com/OpenHands/OpenHands/issues/17233",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.model-picker"+ ],+ "basis": "live-reproduced",+ "reason": "Reproduced the Basic entry dead end described by concrete companion #17577. The authenticated Advanced/save branch was not exercised.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server 1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Reproduced the Basic entry dead end described by concrete companion #17577. The authenticated Advanced/save branch was not exercised.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F10.model-picker/17577-off-catalog.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17213,+ "title": "Rootless Podman without subuid/subgid cannot build OpenHands runtime image",+ "url": "https://github.com/OpenHands/OpenHands/issues/17213",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "reason": "The report is OpenHands 0.54.0 runtime-image construction through ToolShield/rootless Podman, including its legacy runtime Docker template; those runtime-builder paths and UI are absent from this Agent Canvas frontend/Agent Server launcher checkout.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17198,+ "title": "[Bug]: Agent can't access terminal / file system properly: [Command finished execution with no output]",+ "url": "https://github.com/OpenHands/OpenHands/issues/17198",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "basis": "known-issue",+ "reason": "Affected Windows desktop installation, project definitions and a deterministic failing command fixture are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17195,+ "title": "[Bug]: Unable to use 9Router as a custom LLM connection proxying Ollama",+ "url": "https://github.com/OpenHands/OpenHands/issues/17195",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.connection-link",+ "F10.custom-model"+ ],+ "basis": "live-reproduced",+ "reason": "A real Advanced UI profile linked to provider=custom_openai at refused loopback and model=ollama-local/qwen saved successfully, then both actual starts emitted LLM Provider NOT provided before HTTP. The qwen suffix is an owned diagnostic fixture because the report omits its exact exposed model ID; actual 9Router/Ollama inference and that exact ID remain untested.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "steps": [+ "control-openhands api POST /api/llm/provider-connections --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands browser goto /settings/llm",+ "control-openhands browser click 'testid=add-llm-profile'",+ "control-openhands browser click 'testid=sdk-section-advanced-toggle'",+ "control-openhands browser fill 'testid=profile-name-input' QA_ROUTER_MODEL",+ "control-openhands browser fill 'testid=llm-custom-model-input' ollama-local/qwen",+ "control-openhands browser choose 'testid=llm-provider-connection-input' QA_ROUTER_LOOPBACK",+ "control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 1000",+ "control-openhands api GET /api/profiles/QA_ROUTER_MODEL --pick config.model",+ "control-openhands api POST /api/profiles/QA_ROUTER_MODEL/activate --write",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events 1ae226ba-092e-49d9-b55f-b585d66d8fba --kinds ConversationErrorEvent --last 2",+ "control-openhands browser screenshot --feature F10.connection-link --name 17195-ollama-prefix-dispatch",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events 15616fb0-6180-4a07-98ac-470fb455f491 --kinds ConversationErrorEvent --last 2",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write"+ ],+ "evidence": [+ "backfill-18309-settings-rechecks/evidence/F10.connection-link/17195-ollama-prefix-dispatch.png",+ "backfill-18309-settings-rechecks/evidence/F10.connection-link/ollama-prefix-dispatch-diagnostic.json"+ ],+ "observation": "Both starts rejected the reported ollama-local namespace despite a saved custom_openai connection.",+ "fixture_specs": {+ "model": "ollama-local/qwen",+ "connection": {+ "display_name": "QA_ROUTER_LOOPBACK",+ "provider": "custom_openai",+ "base_url": "http://127.0.0.1:1/v1"+ }+ },+ "uncovered_branches": [+ "report does not supply exact model suffix",+ "actual 9Router endpoint backed by Ollama",+ "successful model inference"+ ],+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ },+ "run_id": "backfill-18309-settings-rechecks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17163,+ "title": "[Bug] [Website]: Access denied to model in Cloud (online) - Free account",+ "url": "https://github.com/OpenHands/OpenHands/issues/17163",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.cloud-sandbox-states"+ ],+ "basis": "known-issue",+ "reason": "Affected Free Cloud account, model entitlements and organization history are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17162,+ "title": "[Bug]: /launch trust prompt shows an attacker-controlled source as a GitHub repo coordinate",+ "url": "https://github.com/OpenHands/OpenHands/issues/17162",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F19",+ "feature_ids": [+ "F19.launch-trust"+ ],+ "basis": "live-reproduced",+ "reason": "Crafted evil.example/github.com/...source renders onlyAll-Hands-AI/openhands-plugin and omits evil.example twice; Start remained disabled and no plugin was fetched.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Crafted evil.example/github.com/...source renders onlyAll-Hands-AI/openhands-plugin and omits evil.example twice; Start remained disabled and no plugin was fetched."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F19.launch-trust/17162-source-spoof.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17161,+ "title": "[Bug]: markdown sanitizer lets untrusted content set rel=\"opener\" and any target on links",+ "url": "https://github.com/OpenHands/OpenHands/issues/17161",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.markdown-rendering"+ ],+ "basis": "known-issue",+ "reason": "A real reasoning, activated-skill, tool-description or error-detail body rendered without includeStandard. The attempted skills dialog exposed public skills only and did not reach the affected chat renderer.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17148,+ "title": "test-and-build (windows) reports green while lint, typecheck and the test suite are gated off",+ "url": "https://github.com/OpenHands/OpenHands/issues/17148",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Windows CI lane naming/test gating; repository validation workflow."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17135,+ "title": "Agent Canvas subscription profile editor falls back model and preflight skips subscription auth",+ "url": "https://github.com/OpenHands/OpenHands/issues/17135",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.subscription"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17128,+ "title": "[Bug]: automation menu is cut off and some options are unreachable",+ "url": "https://github.com/OpenHands/OpenHands/issues/17128",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "not-reproduced",+ "family": "F21",+ "feature_ids": [+ "F21.kebab-menu",+ "F21.phone"+ ],+ "reason": "The 25th list row is scrolled into view and its action menu is opened at desktop 1440×1000 and phone 390×844. The full 160×236 menu and its 150×36 Delete button fit within both viewports; no option is cut off in this setup.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use the same 25 inert dashboard records; switch to List via automations-view-toggle and automations-view-toggle-list.",+ "browser click '[data-testid^=\"automation-list-row-\"] >> nth=24 >> role=button[name=\"Automation actions\"]'.",+ "Read bbox 'role=list >> has-text=Export' and the scoped role=button[name=\"Delete\"] at desktop 1440x1000 and phone 390x844.",+ "Observed full 160x236 menu and 150x36 Delete button inside both viewports."+ ],+ "artifacts": [+ "evidence/F21.phone/last-row-kebab-17128.png",+ "evidence/F21.kebab-menu/last-row-phone-17128.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17128#issuecomment-6121792158"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17121,+ "title": "[Bug]: Live run phases never render against automation 1.9.0+ (current_phase)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17121",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.activity-log",+ "F23.run-status-polling"+ ],+ "basis": "live-reproduced",+ "reason": "Actual bounded local script run API current_phase=Starting automation, no structured phase fields; Activity Log reads only Running and contains no phase text.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Live free-form phase (`F23.run-status-polling`, `F23.activity-log`).** Upload an owned bounded script, create it active with the inert yearly cron, open its detail page, and click **Run now**. While genuinely running, compare `api GET /api/automation/v1/<id>/runs` with `browser snapshot 'testid=automation-activity-log'`. Expected: a run's `current_phase` renders as its live/last phase when structured fields are absent. Known failure, reproduced 2026-10-12 against automation 1.19.3: the API says `Starting automation` with no `phase_code`/`phase_label`, but Activity Log shows only **Running** ([#17121](https://github.com/OpenHands/OpenHands/issues/17121)). Keep the worker bounded and clean up only its owned command; the script needs no model."+ ],+ "artifacts": [+ "evidence/F23.activity-log/current-phase-17121.aria.txt",+ "evidence/F23.run-status-polling/missing-phase-17121.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17112,+ "title": "[Bug]: Electron windows can navigate off the loopback app (main window, redirects, nested popups)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17112",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.desktop-main-window",+ "F26.desktop-external-links"+ ],+ "basis": "known-issue",+ "reason": "A runnable owned Electron shell and repeatable popup/redirect fixtures are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17108,+ "title": "[Bug]: The repository picker lists a search result for input the user has already cleared",+ "url": "https://github.com/OpenHands/OpenHands/issues/17108",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.open-repository-dialog"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Cloud Git provider and a real repository search response that remains pending after clearing the input.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17096,+ "title": "docs: DEVELOPMENT.md claims agent-server defaults to latest PyPI",+ "url": "https://github.com/OpenHands/OpenHands/issues/17096",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Agent Server version default documentation drift; documentation only."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17094,+ "title": "docs: stale openhands-dev persistence paths in DEVELOPMENT.md",+ "url": "https://github.com/OpenHands/OpenHands/issues/17094",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Development persistence path documentation drift; documentation only."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17092,+ "title": "docs: stale session-api-key.txt path in AGENTS.md and helm NOTES",+ "url": "https://github.com/OpenHands/OpenHands/issues/17092",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Session API-key filename documentation drift; documentation only."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17069,+ "title": "[Bug]: .env.sample comments disagree with local launchers",+ "url": "https://github.com/OpenHands/OpenHands/issues/17069",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": ".env.sample launcher comments; documentation only, with no runtime change requested."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17061,+ "title": "[Bug]: Renaming a conversation to whitespace only saves an empty title",+ "url": "https://github.com/OpenHands/OpenHands/issues/17061",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.rename-inline"+ ],+ "basis": "live-reproduced",+ "reason": "Whitespace sends PATCH title empty; current server returns422 and restores the prior title.",+ "steps": [+ "On an empty Local conversation: browser dblclick testid=conversation-name-title; fill testid=conversation-name-input \"QA Surface titled\"; browser press Enter; browser reload.",+ "browser network --clear; repeat title edit with one space; browser press Enter.",+ "browser network --filter /api/conversations --bodies => PATCH body title empty, status422.",+ "browser text testid=conversation-name-title and reload => prior title restored."+ ],+ "evidence": [+ "evidence/F07.rename-inline/issue-17061-titled.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17051,+ "title": "[Bug]: Codex work in chat but not in the issue",+ "url": "https://github.com/OpenHands/OpenHands/issues/17051",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "basis": "known-issue",+ "reason": "Codex ACP can work in New Chat while Issue-to-PR automation fails with missing OpenAI API credentials. Reproduction needs an approved Codex subscription login plus a configured Issue-to-PR automation and its repository credentials; the report does not identify the saved automation/profile settings.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17026,+ "title": "agent-canvas: final-reply thinking block missing for text-only turns (reasoning_content not read for MessageEvents)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17026",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.thinking"+ ],+ "basis": "known-issue",+ "reason": "A reasoning-capable live model emitting nonempty MessageEvent.llm_message.reasoning_content on a text-only turn, followed by reload.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16995,+ "title": "[Bug]: Pin as Home is missing from Canvas Extension pages",+ "url": "https://github.com/OpenHands/OpenHands/issues/16995",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F20",+ "feature_ids": [+ "F20.rail-entry"+ ],+ "basis": "live-reproduced",+ "reason": "Installed and enabled reviewed local demo; hover extension sidebar entry reveals no pin control, while hover Automate reveals sidebar-pin-home-toggle-automations.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Extension page home pin (`F20.rail-entry`).** Install and enable the reviewed local `src/fixtures/canvas-extensions/demo-page` through the form and trust confirmation. Hover `sidebar-canvas-extension-demo-page-hello`, inspect `browser testids 'role=complementary' --filter pin`, and compare with hovering `sidebar-automations-link`. Expected: the enabled extension page offers **Pin as home page** and the normal home-pin lifecycle. Known failure, reproduced 2026-10-12 on `fa1f724`: only Automate exposes `sidebar-pin-home-toggle-automations`; the extension page has no pin control ([#16995](https://github.com/OpenHands/OpenHands/issues/16995))."+ ],+ "artifacts": [+ "evidence/F20.rail-entry/extension-no-home-pin-16995.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16988,+ "title": "[Bug]: mock-LLM E2E reports success when the global timeout cut the run short",+ "url": "https://github.com/OpenHands/OpenHands/issues/16988",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Mock-LLM E2E reporter success marker after global timeout; test reporting tooling."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16977,+ "title": "[Bug]: DoneMarkerReporter counts retry attempts as tests, so one flaky retry ends the mock-LLM Docker E2E run early and fails it",+ "url": "https://github.com/OpenHands/OpenHands/issues/16977",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Mock-LLM E2E reporter counts retry attempts; test reporting tooling."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16954,+ "title": "[Bug]: Scope MCP server health by backend and connection revision",+ "url": "https://github.com/OpenHands/OpenHands/issues/16954",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F17",+ "feature_ids": [+ "F17.server-health"+ ],+ "basis": "live-reproduced",+ "reason": "Two SPA switches from a freshly probed Local identity to unprobed QA_SCOPE reused the same Reachable verdict. The identities use ingress/direct URLs of one owned stack; independent credential and delayed-probe branches were not tested.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Two SPA switches from a freshly probed Local identity to unprobed QA_SCOPE reused the same Reachable verdict. The identities use ingress/direct URLs of one owned stack; independent credential and delayed-probe branches were not tested.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F17.server-health/16954-spa-cross-scope.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16951,+ "title": "[Bug]: Centralize effective configuration for the next Home conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/16951",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.composer-blocked-without-llm"+ ],+ "basis": "live-reproduced",+ "reason": "Home readiness follows pinned key presence while implicit launch uses a different active LLM. Both directions observed; actual inverse launch used the keyless active model and omitted named AgentProfile.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server/SDK 1.54.0; automation 1.19.3",+ "viewport": "1440x1000",+ "entry": "Home > named/active LLM mismatch",+ "result": "fail",+ "steps": [+ "Arrange QA_READY key-present and QA_EMPTY keyless LLM profiles with refused loopback URL and num_retries:0",+ "Activate named QA_PINNED agent pointing to QA_EMPTY; activate QA_READY LLM",+ "browser goto /; inspect not-set-up warning and QA_READY pill",+ "Pin QA_PINNED to QA_READY and activate QA_EMPTY; browser goto /",+ "browser fill testid=chat-input <bounded no-tools prompt>; browser enabled testid=submit-button => true",+ "browser click testid=submit-button --expect-url /conversations/",+ "api GET /api/conversations/<id> => active keyless model / launched_agent_profile:null / concurrency1"+ ],+ "actual": "Both readiness/configuration mismatch directions observed; inverse creates real error conversation from keyless active model.",+ "evidence": [+ "evidence/F03.composer-blocked-without-llm/16951-empty-pinned-ready-active.png",+ "evidence/F03.composer-blocked-without-llm/16951-ready-pinned-empty-active.png",+ "evidence/F03.composer-blocked-without-llm/16951-launch-readback.json"+ ],+ "run_id": "backfill-18309-coordinator-checks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16950,+ "title": "[Bug]: Scope Home launch targets and recent repositories by backend and organization",+ "url": "https://github.com/OpenHands/OpenHands/issues/16950",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.selection-after-reload"+ ],+ "basis": "live-reproduced",+ "reason": "Selected secondary workspace remains visible after SPA switch and is submitted to the first Local backend.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "Two run-owned Local Agent Server/SDK1.54.0 instances; automation1.19.3",+ "viewport": "1440x1000",+ "entry": "Home workspace picker and implicit launch across Local backends",+ "result": "fail",+ "steps": [+ "Arrange distinct QA_A_ONLY and QA_B_ONLY workspace registries via each owned /api/workspaces",+ "Register secondary Local backend through real Add Backend UI with its private key file",+ "16950: select secondary QA_B_ONLY and Confirm; switch to Local without reload; inspect preview; submit bounded prompt; actual workspace readback is secondary QA_B_ONLY path",+ "16844: warm Local QA_A_ONLY picker; close dialog; switch to secondary through backend-selector; reopen within60seconds; picker still QA_A_ONLY; real secondary registry remains QA_B_ONLY",+ "Remove registered secondary/workspaces; restore active profile; stop both owned runs"+ ],+ "actual": "Selected secondary workspace remains visible after SPA switch and is submitted to the first Local backend. Cloud organization variant, Windows and isolated same-parent directories remain unverified.",+ "evidence": [+ "evidence/F03.selection-after-reload/16950-target-after-switch.png",+ "evidence/F03.selection-after-reload/16950-submission-readback.json"+ ],+ "run_id": "backfill-18309-coordinator-checks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16949,+ "title": "[Bug]: Keep Files and Changes views source-driven after refresh",+ "url": "https://github.com/OpenHands/OpenHands/issues/16949",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-refresh"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Cloud repository for the refresh-query mismatch, or real file-editor observations causing an in-place M-to-D Git refetch.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16948,+ "title": "[Bug]: Preserve unsaved SDK settings edits across background refetches",+ "url": "https://github.com/OpenHands/OpenHands/issues/16948",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F15",+ "feature_ids": [+ "F15.save-dirty-state"+ ],+ "basis": "live-reproduced",+ "reason": "Reproduced twice on a healthy frozen stack: an unsaved Condenser Max Size=111 was replaced by the changed server baseline=300 and Save disabled after a real stale-query background refetch. No UI Save, reset or scope change occurred. Clock/online state and Max Size=240 restored.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server 1.54.0",+ "viewport": "1440x1000",+ "entry": "Settings > Condenser > All, stale-query reconnect",+ "result": "fail",+ "steps": [+ "browser goto /settings/condenser",+ "browser click testid=sdk-section-all-toggle",+ "browser fill testid=sdk-settings-condenser.max_size 111",+ "api PATCH /api/settings --write --data {agent_settings_diff:{condenser:{max_size:300}}}",+ "browser clock --offset-ms 360000",+ "browser network-state offline",+ "browser network-state online",+ "browser value testid=sdk-settings-condenser.max_size => 300",+ "browser enabled testid=save-button => false"+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F15.save-dirty-state/16948-before-refetch.png",+ "backfill-18309-settings-worker/evidence/F15.save-dirty-state/16948-after-refetch.png",+ "backfill-18309-settings-worker/evidence/F15.save-dirty-state/16948-repeat.png",+ "backfill-18309-settings-worker/evidence/F15.save-dirty-state/network-state-and-refetch.mp4",+ "backfill-18309-settings-worker/evidence/F15.save-dirty-state/network-state-and-refetch.gif"+ ],+ "passes": 2,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16947,+ "title": "[Bug]: Make UI event projection order-independent across history pages",+ "url": "https://github.com/OpenHands/OpenHands/issues/16947",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.load-older-history"+ ],+ "basis": "known-issue",+ "reason": "A real long conversation whose REST boundary separates one action from its observation; retain the separate companion #17559.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16927,+ "title": "[Bug]: Launcher stays alive after a required service dies with EADDRINUSE",+ "url": "https://github.com/OpenHands/OpenHands/issues/16927",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "live-reproduced",+ "reason": "Killed only the owned required static frontend after a healthy startup. The genuine launcher remains alive and authenticated /api/settings still returns 200; UI / returns 502 Bad Gateway. This exercises the required-child fatal-exit branch using SIGKILL, not an EADDRINUSE race; exact trigger is recorded. Normal stack restored afterward.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Required frontend death (`F26.runtime-services`).** After a healthy private launch, identify only its required static-server PID using launcher ancestry plus its owned frontend port, then send that child SIGKILL. Open `/` and check the owned launcher's liveness and authenticated `/api/settings`. Expected: an unexpected required-service death stops the stack and reports failure. Known failure, reproduced 2026-10-12: the launcher and backend remain alive, settings returns 200, and the main UI returns 502 **Bad Gateway** ([#16927](https://github.com/OpenHands/OpenHands/issues/16927)). This live trigger is SIGKILL, not a port-race EADDRINUSE; the failing required-child exit branch is shared. Restart and doctor to restore the run."+ ],+ "artifacts": [+ "evidence/F26.runtime-services/required-frontend-crash-16927.json",+ "evidence/F26.runtime-services/required-frontend-crash-16927.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16905,+ "title": "[Bug]: Claude Code via ACP can't see built-in skills",+ "url": "https://github.com/OpenHands/OpenHands/issues/16905",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "basis": "known-issue",+ "reason": "An enabled built-in OpenHands skill can be absent from a new Claude Code ACP session. The enabled/disabled instruction-delivery check needs an approved authenticated Claude Code account, its ACP subprocess and the configured Code Review automation; no provider session is authorized here.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16893,+ "title": "[Bug]: Skills list is reused after switching registered backends",+ "url": "https://github.com/OpenHands/OpenHands/issues/16893",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F18",+ "feature_ids": [+ "F18.page"+ ],+ "basis": "live-reproduced",+ "reason": "Two SPA backend switches made zero skills requests and omitted a newly arranged card. Reload then fetched actual data and showed it. Registry identities share one physical stack, with real fixture data changes between loads.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Two SPA backend switches made zero skills requests and omitted a newly arranged card. Reload then fetched actual data and showed it. Registry identities share one physical stack, with real fixture data changes between loads.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F18.page/16893-spa-stale-cache.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16877,+ "title": "[Bug]: Untitled terminal-status conversations fast-poll every 3 seconds indefinitely — no give-up for autotitle that never lands",+ "url": "https://github.com/OpenHands/OpenHands/issues/16877",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.open-by-url"+ ],+ "basis": "known-issue",+ "reason": "Real Local idle/null-title check used 30-second polls. The Cloud null-title IDLE/FINISHED branch needs an authenticated Cloud fixture with failed/unsupported autotitle.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server/SDK1.54.0; automation1.19.3",+ "viewport": "1440x1000",+ "entry": "Conversation route > empty idle conversation",+ "result": "pass",+ "steps": [+ "api POST /api/conversations --write --data @<idle-autotitle-disabled-payload> (tools:[], dummy loopback LLM, owned workspace, no initial messages)",+ "browser goto /conversations/<id>",+ "browser network --clear",+ "Leave foreground tab mounted; browser network --filter ids= --last 100",+ "conversation status <id> => title null / idle"+ ],+ "actual": "Eight batch GETs over 210 seconds with 30.008–30.031-second intervals; no 3-second polls. Cloud remains unverified.",+ "evidence": [+ "evidence/F07.open-by-url/16877-local-polling.json",+ "evidence/F07.open-by-url/16877-local-idle.png"+ ],+ "run_id": "backfill-18309-coordinator-checks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16876,+ "title": "[Bug]: Planning-agent history flag stays true forever when sub-conversations fetch fails — infinite chat loading skeleton",+ "url": "https://github.com/OpenHands/OpenHands/issues/16876",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.history-skeleton"+ ],+ "basis": "known-issue",+ "reason": "A real planning sub-conversation whose batch lookup fails or yields null, with the main history available.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16874,+ "title": "[Bug]: Failed cloud sandbox resume is never retried — one-shot ref guard is claimed before the await and cleared on no path",+ "url": "https://github.com/OpenHands/OpenHands/issues/16874",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.cloud-sandbox-states"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Cloud account, paused sandbox and controlled single-request failure are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16872,+ "title": "[Bug]: Electron main process opens arbitrary URI schemes via shell.openExternal without a protocol allowlist",+ "url": "https://github.com/OpenHands/OpenHands/issues/16872",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.desktop-external-links"+ ],+ "basis": "known-issue",+ "reason": "A runnable owned Electron shell and inert OAuth authorization URI fixture are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16871,+ "title": "[Bug]: Reverse tabnabbing in conversation card's VS Code action — window.open without noopener plus scheme-agnostic URL transformer",+ "url": "https://github.com/OpenHands/OpenHands/issues/16871",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.vscode-link"+ ],+ "basis": "known-issue",+ "reason": "An authorized Cloud account/sandbox with a controlled editor exposed_urls value for the current shared-transformer scheme-rejection path. The original card action is absent and the current drawer uses noopener,noreferrer; no exploit or Cloud scheme outcome was reproduced.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F04.download/vscode-card-action-absent-16871.png",+ "backfill-18309-surface-rechecks/evidence/F04.download/vscode-card-action-absent-16871.aria.txt"+ ],+ "source_evidence": [+ "src/components/features/conversation-panel/conversation-card/conversation-card-actions.tsx:131",+ "src/components/features/conversation-panel/conversation-panel.tsx:934",+ "src/hooks/query/use-unified-vscode-url.ts:132",+ "src/components/features/conversation/conversation-tabs/drawer-vscode-link.tsx:25"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16851,+ "title": "[Bug]: wrong model is displayed when loading existing conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/16851",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.profile-identity"+ ],+ "basis": "known-issue",+ "reason": "The original conversation/model-switch history and usage record showing Opus with a Fable selector; a dummy-key conversation has no usage evidence.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16844,+ "title": "[Bug]: Workspace lists are reused after switching registered backends",+ "url": "https://github.com/OpenHands/OpenHands/issues/16844",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.workspace-dropdown-search"+ ],+ "basis": "live-reproduced",+ "reason": "Within2.7seconds of SPA backend switch, saved workspace picker reuses the prior Local registry instead of the active secondary registry.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "Two run-owned Local Agent Server/SDK1.54.0 instances; automation1.19.3",+ "viewport": "1440x1000",+ "entry": "Home workspace picker and implicit launch across Local backends",+ "result": "fail",+ "steps": [+ "Arrange distinct QA_A_ONLY and QA_B_ONLY workspace registries via each owned /api/workspaces",+ "Register secondary Local backend through real Add Backend UI with its private key file",+ "16950: select secondary QA_B_ONLY and Confirm; switch to Local without reload; inspect preview; submit bounded prompt; actual workspace readback is secondary QA_B_ONLY path",+ "16844: warm Local QA_A_ONLY picker; close dialog; switch to secondary through backend-selector; reopen within60seconds; picker still QA_A_ONLY; real secondary registry remains QA_B_ONLY",+ "Remove registered secondary/workspaces; restore active profile; stop both owned runs"+ ],+ "actual": "Within2.7seconds of SPA backend switch, saved workspace picker reuses the prior Local registry instead of the active secondary registry. Cloud organization variant, Windows and isolated same-parent directories remain unverified.",+ "evidence": [+ "evidence/F03.workspace-dropdown-search/16844-local-a-warm.aria.txt",+ "evidence/F03.workspace-dropdown-search/16844-local-b-within-cache.aria.txt",+ "evidence/F03.workspace-dropdown-search/16844-local-b-picker.png",+ "evidence/F03.workspace-dropdown-search/16844-local-switch.mp4",+ "evidence/F03.workspace-dropdown-search/16844-saved-registry-readback.json"+ ],+ "run_id": "backfill-18309-coordinator-checks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16842,+ "title": "[Bug]: A late close from a replaced WebSocket marks the active connection disconnected",+ "url": "https://github.com/OpenHands/OpenHands/issues/16842",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.reconnect"+ ],+ "basis": "known-issue",+ "reason": "A transport fixture delaying an old socket close until its replacement is open; ordinary server restart cannot select that ordering.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16839,+ "title": "[Bug]: Proxy timeout listeners accumulate on keep-alive connections",+ "url": "https://github.com/OpenHands/OpenHands/issues/16839",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "live-reproduced",+ "reason": "Twenty sequential genuine ingress authenticated /api/conversations/search requests over one keep-alive connection, allHTTP 200; the owned ingress emitted MaxListenersExceededWarning for11 timeout listeners.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Keep-alive proxy timeout listeners (`F26.runtime-services`).** Through the owned ingress, send 20 sequential authenticated `/api/conversations/search?limit=1` requests on one HTTP keep-alive connection, recording response statuses and the reused local socket port, then inspect only the owned ingress's warning line. Expected: one socket does not acquire one timeout listener per request. Known failure, reproduced 2026-10-12: all 20 requests are 200 on the same socket and ingress emits `MaxListenersExceededWarning` for 11 timeout listeners ([#16839](https://github.com/OpenHands/OpenHands/issues/16839)). `/server_info` uses a custom enrichment handler here, so it is not the generic proxy branch. Keep header values and full logs private."+ ],+ "artifacts": [+ "evidence/F26.runtime-services/keepalive-16839.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16835,+ "title": "[Bug]: Agent Canvas sidebar shows \"Cache Hit: 0\" while backend correctly records nonzero cache_read_tokens",+ "url": "https://github.com/OpenHands/OpenHands/issues/16835",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "basis": "known-issue",+ "reason": "A working prompt-caching model/proxy and a real conversation reporting nonzero cache_read_tokens.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16802,+ "title": "[Bug]: Codeblocks in markdown \"rich\" render are ugly",+ "url": "https://github.com/OpenHands/OpenHands/issues/16802",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-rich-plain"+ ],+ "basis": "live-reproduced",+ "reason": "Rich fenced code renders nested dark/gray containers and padding; layout-only inspection.",+ "steps": [+ "Arrange <workspace>/QA_rich.md containing a heading and fenced Python print(\"qa\") block.",+ "Open conversation > right-panel-toggle > conversation-tab-files > QA_rich.md.",+ "browser screenshot --feature F08.files-rich-plain; inspect rendered image => nested dark/gray code containers.",+ "Read-only computed styles => pre padding8px12px, outer prose padding24px, borders0px. Layout-only result; no provider call."+ ],+ "evidence": [+ "evidence/F08.files-rich-plain/issue-16802-codeblock.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16801,+ "title": "[Bug]: Agent starts by looking in workspace, even when started with \"no workspace\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/16801",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.launch-without-workspace"+ ],+ "basis": "known-issue",+ "reason": "A working Claude Opus agent and the original task/prompt that makes it explore the default workspace; the prompt is absent.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16774,+ "title": "[Bug]: Mac Installer Produces Damaged Binary",+ "url": "https://github.com/OpenHands/OpenHands/issues/16774",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.desktop-boot-splash",+ "F26.desktop-main-window"+ ],+ "basis": "known-issue",+ "reason": "The reported signed/notarized/quarantined macOS release DMG and an owned installation fixture are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16752,+ "title": "[Bug]: Idle conversations show up the same as actively working conversations",+ "url": "https://github.com/OpenHands/OpenHands/issues/16752",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.card-live-status"+ ],+ "basis": "live-reproduced",+ "reason": "An empty IDLE conversation uses a green active dot; no pulse animation was observed.",+ "steps": [+ "llm set --profile QA_surface --model openai/gpt-4o --base-url http://127.0.0.1:9 --api-key-env QA_DUMMY_KEY --no-validate (dummy value only).",+ "fixture git-repo --name qa-surface-repo; workspace open <run>/workspace/qa-surface-repo.",+ "Sidebar folder-plus > launch-workspace qa-surface-repo, without sending a message.",+ "conversation status <id> => idle, title null.",+ "Read-only browser eval of conversation-status elements => conversation-status-active, green dot, no animation class."+ ],+ "evidence": [+ "evidence/F04.card-live-status/issue-16752-empty.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16750,+ "title": "[Bug]: Conversation list does not load while automation is running",+ "url": "https://github.com/OpenHands/OpenHands/issues/16750",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "basis": "known-issue",+ "reason": "A prompt automation or native delegated task that remains running during conversation-list reload; no usable model is configured here.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16748,+ "title": "[Bug]: CI red on main while every test passes — a late mock XHR throws ProgressEvent ReferenceError",+ "url": "https://github.com/OpenHands/OpenHands/issues/16748",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "MSW/jsdom teardown unhandled rejection; test runner lifecycle."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16735,+ "title": "[Bug]: Plugin launch payload corrupts Unicode parameter values",+ "url": "https://github.com/OpenHands/OpenHands/issues/16735",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F19",+ "feature_ids": [+ "F19.launch-unicode"+ ],+ "basis": "live-reproduced",+ "reason": "A real UTF-8 base64 launch path corrupted Chinese/emoji parameter intoLatin1mojibake twice, including reload. This tests decoding; no source-builder function evaluation was counted asUIproof.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "A real UTF-8 base64 launch path corrupted Chinese/emoji parameter intoLatin1mojibake twice, including reload. This tests decoding; no source-builder function evaluation was counted asUIproof."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F19.launch-unicode/16735-unicode-parameters.png",+ "backfill-18309-settings-worker/evidence/F19.launch-unicode/16735-unicode-parameters.aria.txt"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16727,+ "title": "[Bug]: Conversation corrupted after VM restart: \"Missing event file\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/16727",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.open-by-url"+ ],+ "basis": "known-issue",+ "reason": "A disposable VM interrupted at the event-write boundary or the original missing-event persisted fixture.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16726,+ "title": "[Bug]: Missing BaseUrl for adding LLM Profile for Azure provider",+ "url": "https://github.com/OpenHands/OpenHands/issues/16726",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.model-picker"+ ],+ "basis": "live-reproduced",+ "reason": "Azure Basic had no Base URL input; Advanced had one, and returning to Basic hid it again. No authenticated Azure conversation was attempted.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "local Agent Server 1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Azure Basic had no Base URL input; Advanced had one, and returning to Basic hid it again. No authenticated Azure conversation was attempted.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F10.model-picker/16726-azure-basic.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16724,+ "title": "[Bug]: vercel-install.sh overwrites its first insteadOf rule, so the ssh:// rewrite never applies",+ "url": "https://github.com/OpenHands/OpenHands/issues/16724",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Vercel dependency-install Git rewrite configuration; deployment build tooling with no current Canvas runtime reproduction."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16720,+ "title": "[Bug]: Agent commands should run with lower priority (higher number)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16720",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "known-issue",+ "reason": "A bounded reproducible CPU-heavy workload, affected resource envelope and model key are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16719,+ "title": "[Bug]: Agent status always says \"thinking\" even when nothing is happening",+ "url": "https://github.com/OpenHands/OpenHands/issues/16719",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.live-activity"+ ],+ "basis": "known-issue",+ "reason": "The event/status trace and deterministic stalled or errored state from the original report; its trigger is unspecified.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16716,+ "title": "[Bug]: Diff view is regularly out of sync",+ "url": "https://github.com/OpenHands/OpenHands/issues/16716",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "basis": "known-issue",+ "reason": "The original simultaneous-agent/shared-repository state or a deterministic stale-diff trigger; the report only suggests that timing.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16690,+ "title": "Agent Canvas EXTREMELY slow and unusable >3 conversations and ECONN REFUSED error",+ "url": "https://github.com/OpenHands/OpenHands/issues/16690",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.docker-image",+ "F26.runtime-services"+ ],+ "basis": "known-issue",+ "reason": "Affected Windows-host Docker setup and persisted conversation/workload fixture are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16689,+ "title": "[Bug]: When `workspace/` paths throw a 401, call the workspace-session endpoint",+ "url": "https://github.com/OpenHands/OpenHands/issues/16689",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-load-error"+ ],+ "basis": "known-issue",+ "reason": "The original fresh-VM URL, auth/cookie state and exact request sequence; the report says the trigger is unclear.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16686,+ "title": "[Bug]: LLM errors not propagated to the frontend",+ "url": "https://github.com/OpenHands/OpenHands/issues/16686",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "A Nebius credential and its rejected DeepSeek-V3 request, or the original complete emitted error fixture.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16611,+ "title": "Plugin-preset automations fail with 500 on agent-server conversation/events path (prompt presets work)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16611",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F22",+ "feature_ids": [+ "F22.setup-plugin-create"+ ],+ "basis": "known-issue",+ "reason": "Authenticated Cloud account, model key and reviewed no-op plugin fixture are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16601,+ "title": "[Bug]: Analytics consent reappears sporadically",+ "url": "https://github.com/OpenHands/OpenHands/issues/16601",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F01",+ "feature_ids": [+ "F01.telemetry-consent"+ ],+ "basis": "known-issue",+ "reason": "The multi-day trigger and backend consent response when the modal returns; opted-in localStorage alone does not reproduce it.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16583,+ "title": "[Bug]: Readiness and PR-description checks count fenced example text as real content",+ "url": "https://github.com/OpenHands/OpenHands/issues/16583",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Issue-readiness and PR-description validation of fenced examples; repository tooling."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16582,+ "title": "use-handle-ws-events: send missing from useEffect deps — stale WebSocket closure after reconnect silently drops AgentState.PAUSED message",+ "url": "https://github.com/OpenHands/OpenHands/issues/16582",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "basis": "known-issue",+ "reason": "A maximum-iteration legacy error delivered after socket replacement, with a real model run reaching that limit.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16581,+ "title": "use-terminal: persistentLastCommandIndex is a module-level singleton — concurrent Terminal instances share one counter, second terminal renders no commands",+ "url": "https://github.com/OpenHands/OpenHands/issues/16581",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "basis": "known-issue",+ "reason": "A supported host fixture with simultaneous independent Terminal consumers and real command histories is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16580,+ "title": "use-drag-resize: touchmove/touchend listeners registered with capture:true but removed without it — permanent listener accumulation on mobile after each drag",+ "url": "https://github.com/OpenHands/OpenHands/issues/16580",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.composer-resize"+ ],+ "basis": "known-issue",+ "reason": "Real touch drags or a CLI touch-gesture verb. Current cleanup already removes capture listeners with capture=true; this CLI exposes mouse drag only.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16579,+ "title": "use-load-older-events: stale conversationId closure after await injects wrong conversation events and permanently disables load-older for new conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/16579",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.load-older-history"+ ],+ "basis": "known-issue",+ "reason": "A delayed older-history response released after another conversation loads; current source has an active-load guard, but the exact race needs a transport fixture.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16578,+ "title": "[Bug]: litellm.APIConnectionError: No module named 'boto3'",+ "url": "https://github.com/OpenHands/OpenHands/issues/16578",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "family": "F10",+ "feature_ids": [+ "F10.default-used",+ "F10.custom-model"+ ],+ "basis": "live-not-reproduced",+ "reason": "Two actual Canvas Bedrock starts reached boto3.Session and the expected qa_nonexistent profile-not-found failure; their real traces did not contain No module named boto3. AWS metadata was disabled in the owned server environment, and both runtime/base endpoints were refused loopback. This establishes current Local SDK1.54.0 named-profile import, not valid AWS credentials, successful inference, historical Linux dependency resolution or fresh-install variants.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "steps": [+ "python3 - <<'PY'\nimport json,shutil\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(p,backup);d=json.loads(p.read_text());d['AWS_EC2_METADATA_DISABLED']='true';p.write_text(json.dumps(d));print({'AWS_EC2_METADATA_DISABLED':True,'run_owned_launcher_env_backed_up':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands api POST /api/profiles/QA_BEDROCK --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands api POST /api/profiles/QA_BEDROCK/activate --write",+ "control-openhands doctor",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events f8b5b78c-3079-4049-b902-9d4a065b0ce9 --kinds ConversationErrorEvent --last 2",+ "control-openhands browser screenshot --feature F10.default-drives-conversation --name 16578-boto3-import-diagnostic",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events f3021128-a9cc-4d31-b6e7-a8434894871c --kinds ConversationErrorEvent --full --last 1 > 'backfill-18309-settings-rechecks/<private-fixture>'",+ "control-openhands conversation events f8b5b78c-3079-4049-b902-9d4a065b0ce9 --kinds ConversationErrorEvent --full --last 1 > backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands doctor"+ ],+ "evidence": [+ "backfill-18309-settings-rechecks/evidence/F10.default-drives-conversation/bedrock-import-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F10.default-drives-conversation/16578-boto3-import-diagnostic.png"+ ],+ "observation": "Both starts imported boto3 and failed at missing named AWS profile before any AWS/provider request.",+ "fixture_specs": {+ "llm": {+ "model": "bedrock/anthropic.claude-3-haiku-20240307-v1:0",+ "base_url": "http://127.0.0.1:1/v1",+ "aws_bedrock_runtime_endpoint": "http://127.0.0.1:1/v1",+ "aws_region_name": "us-east-1",+ "aws_profile_name": "qa_nonexistent",+ "num_retries": 0+ },+ "include_secrets": true+ },+ "setup_mistakes": [+ "Initial keyless fixture hit Canvas launch gate (llm_api_key_is_set=false), so no conversation was created; excluded from proof. Corrected dummy-only presence marker allowed the SDK Bedrock named-profile initialization."+ ],+ "uncovered_branches": [+ "valid AWS credentials/Bedrock permissions",+ "successful inference",+ "historical Linux npm dependency resolution",+ "fresh uncached dependency-resolution variants"+ ],+ "replay_setup": {+ "working_directory": "<checkout>",+ "run": "Export OH_VERIFY_RUN as a fresh owned run directory; substitute actual conversation ids and private fixture paths.",+ "launcher_preconditions": {+ "AWS_EC2_METADATA_DISABLED": "true",+ "OPENAI_BASE_URL": "http://127.0.0.1:1/v1",+ "OPENAI_API_BASE": "http://127.0.0.1:1/v1"+ },+ "verification": "Set only run-private launcher environment, restart and doctor; confirm only booleans for the three flags on the process bound to the owned Agent Server port. No credentials or full environment output.",+ "profile_body": {+ "llm": {+ "model": "bedrock/anthropic.claude-3-haiku-20240307-v1:0",+ "base_url": "http://127.0.0.1:1/v1",+ "aws_bedrock_runtime_endpoint": "http://127.0.0.1:1/v1",+ "aws_region_name": "us-east-1",+ "aws_profile_name": "qa_nonexistent",+ "num_retries": 0+ },+ "include_secrets": true+ },+ "scope": "Named missing profile must fail before AWS/provider HTTP; both runtime endpoints are explicitly loopback. Dummy key is only for Canvas launch gate; no actual AWS credentials."+ },+ "replay_notes": [+ "Use existing stable F10.default-used in new screenshot/evidence commands. Original artifact files remain under their captured folder path, whose first evidence tag was corrected."+ ],+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ },+ "run_id": "backfill-18309-settings-rechecks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/16578#issuecomment-6122466700"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16551,+ "title": "[Bug]: [Windows] Browser tool prioritizes system Chrome over bundled Playwright Chromium, causing silent CDP hang in headless mode",+ "url": "https://github.com/OpenHands/OpenHands/issues/16551",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.browser-screenshot"+ ],+ "basis": "known-issue",+ "reason": "Native Windows with both system Chrome and Playwright Chromium plus a model key is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16530,+ "title": "IMP: Security issue is disclosed but still not fixed",+ "url": "https://github.com/OpenHands/OpenHands/issues/16530",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "reason": "This is a request to disclose an unspecified security report and request a CVE; it names no Canvas behavior, affected feature, vulnerability or reproducible failure, so there is no feature-map surface to classify.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16526,+ "title": "403 error when selecting an agent in the onboarding agent selection modal",+ "url": "https://github.com/OpenHands/OpenHands/issues/16526",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F01",+ "feature_ids": [+ "F01.onboarding-choose-agent"+ ],+ "basis": "known-issue",+ "reason": "An authenticated hosted Cloud account with the reported organization-membership mismatch during onboarding.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16511,+ "title": "[Bug]: Docker quickstart fails: bind-mount permission error kills agent-server and automation silently",+ "url": "https://github.com/OpenHands/OpenHands/issues/16511",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.docker-image"+ ],+ "basis": "known-issue",+ "reason": "An owned Docker daemon/image and disposable mount ownership fixture are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16506,+ "title": "[Bug]: Flaky CI failure — ReferenceError: ProgressEvent is not defined",+ "url": "https://github.com/OpenHands/OpenHands/issues/16506",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "MSW/jsdom ProgressEvent test teardown race; test runner lifecycle."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16492,+ "title": "Backend API keys live in origin-readable localStorage — any same-origin script can read all of them",+ "url": "https://github.com/OpenHands/OpenHands/issues/16492",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.manage-backends",+ "F25.add-agent-server"+ ],+ "basis": "live-reproduced",+ "reason": "Read-only same-origin browser inspection records the number of registered backend entries and nonempty plaintext apiKey values; the retained probe now shows both owned entries are readable. No credential value or whole storage JSON was printed.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Same-origin credential readability (`F25.manage-backends`, `F25.add-agent-server`).** In an authenticated private run, use read-only `browser eval` to return only the count of registered entries and the count whose `apiKey` is a nonempty string in `JSON.parse(localStorage.getItem('openhands-backends') || '[]')`. Expected: backend credentials are not readable by arbitrary same-origin script. Known failure, reproduced 2026-10-12: registered entries have readable plaintext keys ([#16492](https://github.com/OpenHands/OpenHands/issues/16492)). Retain counts/booleans only, never key values or whole storage JSON."+ ],+ "artifacts": [+ "evidence/F25.manage-backends/plaintext-key-counts-16492.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16455,+ "title": "[Bug]: Duplicate messages from the agent in 1.12.0",+ "url": "https://github.com/OpenHands/OpenHands/issues/16455",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.events-match"+ ],+ "basis": "known-issue",+ "reason": "The original duplicated event/stream trace or deterministic prompt and timing; the report provides no reproducible trigger.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16454,+ "title": "[Bug]: Diff list is showing duplicate entries",+ "url": "https://github.com/OpenHands/OpenHands/issues/16454",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "basis": "known-issue",+ "reason": "The original Git changes response and update sequence producing duplicate rows; the report provides no trigger.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16430,+ "title": "[Bug]: File upload does not have an indication of progress",+ "url": "https://github.com/OpenHands/OpenHands/issues/16430",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.attach-files"+ ],+ "basis": "known-issue",+ "reason": "A remote or throttled real upload that stays in flight long enough to inspect progress and composer state.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16429,+ "title": "fix: grant release-ready contents write permission",+ "url": "https://github.com/OpenHands/OpenHands/issues/16429",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Release-ready workflow permissions; repository CI tooling."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16399,+ "title": "[Bug]: Pinned conversations from later pages are removed after reload",+ "url": "https://github.com/OpenHands/OpenHands/issues/16399",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.pin"+ ],+ "basis": "live-reproduced",+ "reason": "A page-two pin disappears after first-page reload although the conversation still exists.",+ "steps": [+ "Create23 empty conversations via sidebar folder-plus workspace launch, leaving the older QA Surface titled conversation outside the first20.",+ "browser goto /; browser reload; browser click role=button[name=\"Load more\"].",+ "browser hover conversation-card with QA Surface titled; click its conversation-pin-toggle control.",+ "browser text testid=conversation-panel-pinned-section => Pinned, QA Surface titled.",+ "browser reload; browser count testid=conversation-panel-pinned-section =>0; conversation still retained on server."+ ],+ "evidence": [+ "evidence/F04.pin/issue-16399-reload.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16383,+ "title": "Session API key passed as plaintext CLI arg to static-server.mjs (visible via ps/proc)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16383",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "live-reproduced",+ "reason": "Inspect only the owned static-server process for frontend18813: session-key equals a literal argv value. Evidence records booleans only; no credential/argv string exposed.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Static-server credential argv (`F26.runtime-services`).** Inspect only this run's owned static-server process for the frontend port. Read the private key in memory and retain booleans indicating whether `--session-api-key` is present and whether that literal key occurs in argv; never print argv or the key. Expected: other host users cannot recover the session key from command-line arguments. Known failure, reproduced 2026-10-12: both booleans are true for the owned static frontend ([#16383](https://github.com/OpenHands/OpenHands/issues/16383)). Docker remains a separate unavailable prerequisite here."+ ],+ "artifacts": [+ "evidence/F26.runtime-services/argv-16383.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16374,+ "title": "RemoteSandboxService: extend dedicated read-session fix to remaining read paths (follow-up to #14769, finding #1)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16374",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "Cloud app-server SQLAlchemy read-session atomicity has no Canvas surface or client-owned contract; implementation is outside this frontend."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16360,+ "title": "Agent status text overflows vertically on mobile",+ "url": "https://github.com/OpenHands/OpenHands/issues/16360",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "The issue is labeled Legacy Cloud and describes the removed v0 Cloud composer/status layout, not the current Canvas composer."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16356,+ "title": "Conversation LLM config is frozen at creation time — resuming after API key rotation fails permanently with LLMAuthenticationError",+ "url": "https://github.com/OpenHands/OpenHands/issues/16356",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.open-by-url"+ ],+ "basis": "known-issue",+ "reason": "A working provider key that can be rotated/revoked, an existing conversation using it, and a replacement profile.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16341,+ "title": "[Bug]: Code-explorer subagent frequently errors",+ "url": "https://github.com/OpenHands/OpenHands/issues/16341",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A working model with native code-explorer delegation and the original repository task; no delegated-tool output can be generated without credentials.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16339,+ "title": "[Bug] Agent not informed of file location for files uploaded via canvas in cloud",+ "url": "https://github.com/OpenHands/OpenHands/issues/16339",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.attach-files"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Cloud upload sandbox and working model to inspect the delivered filename/path notification and read the uploaded file.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16331,+ "title": "[Bug]: file preview panel stops working during streaming",+ "url": "https://github.com/OpenHands/OpenHands/issues/16331",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-tree"+ ],+ "basis": "known-issue",+ "reason": "A real streaming/tool-running conversation while selecting a second workspace file.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16300,+ "title": "[Bug]: litellm build fails with rustc version mismatch on macOS/Linux",+ "url": "https://github.com/OpenHands/OpenHands/issues/16300",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "known-issue",+ "reason": "An uncached isolated reproduction of the reported LiteLLM/Rust dependency toolchain is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16298,+ "title": "[Bug]: slack-monitor skill is not available in 1.9.0",+ "url": "https://github.com/OpenHands/OpenHands/issues/16298",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F18",+ "feature_ids": [+ "F18.skill-in-chat"+ ],+ "basis": "known-issue",+ "reason": "A Slack-monitor setup can leave the agent without the expected skill. The report omits the concrete monitor setup and loaded skill name/version; reproducing the runtime also needs an approved model and a configured Slack integration. Listing the bundled catalog alone cannot settle it.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16297,+ "title": "[Bug]: Default \"Local\" backend name is confusing when on a VM",+ "url": "https://github.com/OpenHands/OpenHands/issues/16297",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.selector-dropdown"+ ],+ "basis": "live-reproduced",+ "reason": "Served the unchanged UI on an owned protected non-loopback LAN origin, with auth-required and no injected key. Onboarding default Host Name is Local despite the LAN URL; after authenticating with only the private run key and skipping remaining setup, the actual selector reads Connected Local. This verifies the non-loopback naming branch, not a VM installation.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Non-loopback default backend name (`F25.selector-dropdown`).** Serve the unchanged frontend on an owned non-loopback LAN address with `--auth-required`, no session-key injection and real protected backend routes. Open that origin in a new browser tab with `--allow-external`; its Add a backend step defaults **Host Name** to `Local`. Authenticate only with the private run key file, continue once and skip the remaining setup, then inspect `backend-selector`. Expected: the default name identifies the backend's host on a remote origin. Known failure, reproduced 2026-10-12: the selector reads **Connected Local** on the non-loopback origin ([#16297](https://github.com/OpenHands/OpenHands/issues/16297)). This is a LAN-origin check, not a VM deployment claim. Stop only the owned fixture frontend and return to the original tab."+ ],+ "artifacts": [+ "evidence/F25.selector-dropdown/nonloopback-label-final-16297.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16296,+ "title": "[Bug]: Hide claude code config fields if connected",+ "url": "https://github.com/OpenHands/OpenHands/issues/16296",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.acp-credentials"+ ],+ "basis": "known-issue",+ "reason": "Claude Code credential fields remain visible below the already-connected banner, leading users to think keys are required. The connected-state check needs an approved existing Claude CLI login on the agent-server host; a dummy credential does not prove that signed-in state.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16277,+ "title": "Automate tab crashes on catalog entry with missing `requires` (guard getIntegrationIds)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16277",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F22",+ "feature_ids": [+ "F22.template-card",+ "F22.templates-page"+ ],+ "basis": "live-reproduced",+ "reason": "Frozen-source isolated copy with only private @openhands/extensions custom-automation manifest.requires removed; build served correctly from build/ against the healthy owned backend. Actual /automations/templates renders Application Error with TypeError: Cannot read properties of undefined (reading integrations), crashing the entire catalog instead of isolating the malformed entry. Shared source/build untouched.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Malformed catalog entry (`F22.template-card`, `F22.templates-page`).** Archive the frozen target into an isolated private checkout, link unchanged dependencies except a private copy of the pinned `@openhands/extensions`, and remove only `requires` from `automations/catalog/custom-automation/manifest.json`. Leave Canvas source and its lockfile unchanged; run `npm run build:app` only in that copy. Serve its flat `build/` directory against an owned healthy backend, then open `/automations/templates` through `browser goto <fixture-origin>/automations/templates --allow-external`. Expected: the malformed entry degrades without crashing the valid catalog. Known failure, reproduced 2026-10-12 on `fa1f724`: the whole page renders **Application Error**, `TypeError: Cannot read properties of undefined (reading 'integrations')` ([#16277](https://github.com/OpenHands/OpenHands/issues/16277)). Record this as catalog-data fault injection, then stop/remove only the private fixture frontend/build; never modify the shared package or build."+ ],+ "artifacts": [+ "evidence/F22.template-card/missing-requires-16277.png",+ "evidence/F22.template-card/missing-requires-16277.aria.txt"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16263,+ "title": "Fix LLM selector reverting to 'Select a model' for failed conversations",+ "url": "https://github.com/OpenHands/OpenHands/issues/16263",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.profile-identity"+ ],+ "basis": "live-reproduced",+ "reason": "An errored conversation with an unmatched model shows Select a model despite loaded profiles.",+ "steps": [+ "Create a Local conversation using dummy QA_surface profile with openai/gpt-4o and closed loopback base_url.",+ "browser reset; onboard --skip to clear browser profile stamp; reopen conversation.",+ "Send \"Reply with only QA. Do not run any tools.\" from chat-input; closed loopback later returns error.",+ "llm set same QA_surface profile to openai/gpt-4o-mini, same dummy loopback credentials, --no-validate.",+ "browser reload; browser text chat-input-llm-profile => Select a model; conversation status => error, persisted model openai/gpt-4o."+ ],+ "evidence": [+ "evidence/F05.profile-identity/issue-16263-error.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16257,+ "title": "dev launch scripts install signal handlers as import-time side effects",+ "url": "https://github.com/OpenHands/OpenHands/issues/16257",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "reason": "The issue explicitly scopes an import-time signal-handler ownership/dev-tooling refactor and states there is no user-facing feature or child-service behavior change. No observed Canvas startup/shutdown failure is supplied for the map; concrete required-service failures remain mapped separately in #16927/#18299.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16226,+ "title": "[Bug]: FIXED frontend port results in port conflict",+ "url": "https://github.com/OpenHands/OpenHands/issues/16226",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.cli-port-in-use"+ ],+ "basis": "live-reproduced",+ "reason": "Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. The explicit ingress override does not choose another frontend port.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Occupied internal frontend port (`F26.cli-port-in-use`).** Reserve a loopback frontend port in an owned listener and run the genuine launcher with separate free ingress/backend/automation ports. The reported default-port branch can also be observed read-only when port 3001 is already occupied; do not stop its owner. Expected: an occupied internal frontend port receives a free fallback while explicit service-port conflicts keep their existing semantics. Known failure, reproduced 2026-10-12: `agent-canvas --port 18940` exits 1 with `frontend: port 3001`, despite the ingress override ([#17546](https://github.com/OpenHands/OpenHands/issues/17546), [#16226](https://github.com/OpenHands/OpenHands/issues/16226)). An explicit `OH_CANVAS_SAFE_VITE_PORT` is a current workaround; it does not provide automatic fallback."+ ],+ "artifacts": [+ "evidence/F26.cli-port-in-use/frontend-default-17546-16226.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16217,+ "title": "[Bug]: npm run dev can launch Uvicorn outside the automation environment",+ "url": "https://github.com/OpenHands/OpenHands/issues/16217",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "not-reproduced",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "reason": "Installed real uvicorn 0.37.0 alone in an owned Python 3.12 venv, prepended that bin to the private launcher PATH, then restarted the genuine current launcher. Automation 1.19.3 starts and /api/automation/health returns HTTP 200; the reported external-uvicorn ModuleNotFoundError does not occur in this exact resolver/PATH setup. Original PATH restored and doctor run afterward.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "uv venv --python 3.12 <run>/private/uvicorn-only; uv pip install --python <venv>/bin/python uvicorn==0.37.0.",+ "Save <run>/private/launcher-env.json; prepend only <venv>/bin to that owned run's PATH; control-openhands restart --timeout 30.",+ "control-openhands api GET /api/automation/health returned HTTP 200; the genuine current uvx-launched automation starts despite the standalone executable.",+ "Restore the saved launcher PATH, restart, doctor. This is scoped to current automation 1.19.3 and this isolated PATH fixture."+ ],+ "artifacts": [+ "evidence/F26.runtime-services/standalone-uvicorn-16217.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/16217#issuecomment-6121792551"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16210,+ "title": "[Cloud Backends] Plugins not available when starting a new conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/16210",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.plugin-picker"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Cloud organization with configured account plugins before opening New Chat > Plugins.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16208,+ "title": "[Bug]: Agent Canvas 1.7.2: POST /api/conversations returns 500 on Fedora Rawhide (Python 3.14, local npm install)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16208",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "known-issue",+ "reason": "Fedora Rawhide/Python 3.14.6 environment, reported SDK/runtime and model key are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16161,+ "title": "CI stability audit: unpinned/floating dependencies that can flip CI from green to red",+ "url": "https://github.com/OpenHands/OpenHands/issues/16161",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "CI/dependency publication stability audit; contributor build and release tooling."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16155,+ "title": "[Bug]: \"Run now\" button renders incorrectly when the automation title is long",+ "url": "https://github.com/OpenHands/OpenHands/issues/16155",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "not-reproduced",+ "family": "F21",+ "feature_ids": [+ "F21.list-row",+ "F21.run-now"+ ],+ "reason": "The exact reported long title is arranged in a real inactive list row and scrolled into view. Run now is an icon-only 32×32 px control, with no wrapping Run/now text, in the current list layout.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Create an inactive automation named 'GitHub PR Reviewer: OpenHands/software-agent-sdk for @all-hands-bot'.",+ "browser goto /automations; switch to List; browser scroll '[data-testid^=\"automation-list-row-\"] >> has-text=GitHub PR Reviewer'.",+ "Read row snapshot and Run now bbox; take long-title-visible-16155 screenshot.",+ "Observed icon-only Run now control at 32x32 px and a single line title; original wrapping Run/now text is absent in this layout."+ ],+ "artifacts": [+ "evidence/F21.list-row/long-title-visible-16155.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/16155#issuecomment-6121792984"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15421,+ "title": "[Bug]: Configured MCP servers not being picked up by the agent",+ "url": "https://github.com/OpenHands/OpenHands/issues/15421",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F17",+ "feature_ids": [+ "F17.agent-uses-server"+ ],+ "basis": "known-issue",+ "reason": "A configured Linear MCP can be ignored while an explicitly disabled Linear skill is used instead. The new-chat and automation paths need a working authorized Linear MCP connection, a concrete issue ID and an approved model; the report's automation ZIP is also needed to replay its exact automation settings.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15430,+ "title": "OAuth MCP callback fails in Docker: localhost redirect port not exposed",+ "url": "https://github.com/OpenHands/OpenHands/issues/15430",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F17",+ "feature_ids": [+ "F17.install-oauth"+ ],+ "basis": "known-issue",+ "reason": "An OAuth MCP callback in Docker can redirect to an unexposed dynamic localhost port. Completing the callback needs a Docker stack and an approved OAuth MCP provider/account (the report uses GitLab); a local unauthenticated Test connection cannot reach the redirect.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15433,+ "title": "[Bug]: user messages can split streaming text",+ "url": "https://github.com/OpenHands/OpenHands/issues/15433",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.user-and-agent-messages"+ ],+ "basis": "known-issue",+ "reason": "A working streaming model and follow-up sent while its response is still arriving.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15434,+ "title": "Embedded VS Code button unreachable in self-hosted mode — advertised URL uses page origin while the server binds a separate port, with no config to reconcile them",+ "url": "https://github.com/OpenHands/OpenHands/issues/15434",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.remote-backend"+ ],+ "basis": "known-issue",+ "reason": "Affected non-localhost HTTPS reverse proxy deployment and functioning editor binary are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15458,+ "title": "[Bug]: Customize > Skills is showing agent.md files inside of conversation worktrees from automation-runs",+ "url": "https://github.com/OpenHands/OpenHands/issues/15458",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F18",+ "feature_ids": [+ "F18.page",+ "F18.project-skill"+ ],+ "basis": "known-issue",+ "reason": "The global Skills page can list AGENTS.md files from cloned automation-run worktrees. The report says the discovery trigger is unknown; replay needs the affected automation workspace/discovery-root layout or captured request parameters, rather than only a normal project-skill fixture.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15465,+ "title": "[Bug]: Event based Automation doesn't seem to work",+ "url": "https://github.com/OpenHands/OpenHands/issues/15465",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.configuration-event",+ "F23.activity-log"+ ],+ "basis": "known-issue",+ "reason": "Externally reachable responder, authorized webhook/release repository and controlled real event delivery are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15495,+ "title": "[Bug]: Bedrock documentation is lacking",+ "url": "https://github.com/OpenHands/OpenHands/issues/15495",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "known-issue",+ "reason": "The request is for missing AWS Bedrock setup documentation at docs.openhands.dev, with no concrete Canvas control or failing UI behavior. F10 consumes profile forms, not the external documentation authoring surface.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15501,+ "title": "[Bug]: allow-scripts inside iframe",+ "url": "https://github.com/OpenHands/OpenHands/issues/15501",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-rich-plain"+ ],+ "basis": "known-issue",+ "reason": "The original preview panel and file URL or minimal page identifying the intended surface; Files Rich deliberately blocks scripts, while Browser is separate.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15509,+ "title": "[Bug]: Automation fails on Windows local-mode: \"Path must be absolute\" during tarball upload",+ "url": "https://github.com/OpenHands/OpenHands/issues/15509",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F22",+ "feature_ids": [+ "F22.setup-tarball-create"+ ],+ "basis": "known-issue",+ "reason": "Native Windows stack and controlled event/webhook delivery fixture are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15519,+ "title": "[Bug]: \"files\" tab in Files side panel throws 401 Unauthorized",+ "url": "https://github.com/OpenHands/OpenHands/issues/15519",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-load-error"+ ],+ "basis": "known-issue",+ "reason": "The reported Linux Docker deployment/auth topology; this isolated run uses the macOS source launcher.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15527,+ "title": "[Bug]: Canvas briefly shows \"disconnected\" when coming back to an idle tab",+ "url": "https://github.com/OpenHands/OpenHands/issues/15527",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.reconnect"+ ],+ "basis": "known-issue",+ "reason": "The original custom-domain deployment, idle duration and tab-suspension/network transition; those details are missing.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15543,+ "title": "[Bug]: Bad parsing for branch retrieval",+ "url": "https://github.com/OpenHands/OpenHands/issues/15543",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.git-control-bar"+ ],+ "basis": "known-issue",+ "reason": "The original Git probe stdout/interleaving that temporarily puts a filename in the branch badge; dirty state alone is not a deterministic trigger.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15567,+ "title": "Workspace file preview returns 401 Unauthorized on non-HTTPS non-loopback hosts",+ "url": "https://github.com/OpenHands/OpenHands/issues/15567",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-load-error"+ ],+ "basis": "known-issue",+ "reason": "A plain-HTTP non-loopback deployment and browser accessing that origin, with workspace-cookie observation.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15599,+ "title": "Agent creates redundant refs when working with PRs instead of using the PR source branch",+ "url": "https://github.com/OpenHands/OpenHands/issues/15599",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "The report concerns an agent choosing a Git ref strategy from a free-form task; it identifies no Canvas state, request or Git-control defect."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15606,+ "title": "[Bug]: When re-loading ACP conversations, messages disappear or move",+ "url": "https://github.com/OpenHands/OpenHands/issues/15606",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.reload-mid-run"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Codex ACP session emitting intermediate messages while another window reloads history.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15626,+ "title": "[Bug]: Conversation can fail with LiteLLM auth error when database server at 127.0.0.1:5432 is unreachable",+ "url": "https://github.com/OpenHands/OpenHands/issues/15626",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "The original LiteLLM proxy/database-failure event or a disposable proxy with its PostgreSQL dependency unavailable.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15662,+ "title": "[Bug]: Cannot select existing conversation from collapsed side nav bar",+ "url": "https://github.com/OpenHands/OpenHands/issues/15662",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F02",+ "feature_ids": [+ "F02.sidebar-collapse"+ ],+ "basis": "live-reproduced",+ "reason": "Collapsed rail has no existing-conversation chooser; expansion is required.",+ "steps": [+ "Arrange an existing empty Local conversation via workspace open and sidebar folder-plus > launch-workspace.",+ "browser click testid=sidebar-collapse-toggle",+ "browser snapshot => Search commands, New Chat, Customize, Automate, Settings, Manage Backends; no conversation chooser.",+ "browser count testid=conversation-panel => 0.",+ "Restore: browser hover role=link[name=\"OpenHands Logo\"], then click testid=sidebar-collapse-toggle."+ ],+ "evidence": [+ "evidence/F02.sidebar-collapse/issue-15662-collapsed.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F02-app-shell.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15676,+ "title": "[Bug]: Refreshing stale Cloud conversation on Vercel fails",+ "url": "https://github.com/OpenHands/OpenHands/issues/15676",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.cloud-sandbox-states"+ ],+ "basis": "known-issue",+ "reason": "Affected Vercel Cloud session, aged sandbox and exact error/deployment revision are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15692,+ "title": "[Bug]: Diff still shows after agent has committed and pushed",+ "url": "https://github.com/OpenHands/OpenHands/issues/15692",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "basis": "known-issue",+ "reason": "A real agent commit/push terminal observation while the diff remains mounted, plus a disposable Git remote.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15694,+ "title": "Bug: Enable/disable skills toggle resets to enabled state unexpectedly",+ "url": "https://github.com/OpenHands/OpenHands/issues/15694",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "family": "F18",+ "feature_ids": [+ "F18.toggle"+ ],+ "basis": "live-not-reproduced",+ "reason": "The disabled skill stayed off in two eight-second samples and reloads; the server retained its disabled name. The report is intermittent, so longer or different timing remains unverified.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands fixture skill --name qa-toggle --trigger qa-toggle-ping",+ "control-openhands browser goto /skills",+ "control-openhands browser reload",+ "control-openhands browser fill 'testid=skills-search-input' qa-toggle",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250 --observe 'testid=skill-toggle-qa-toggle' --observe-ms 8000",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands api GET /api/settings --pick misc_settings.app_preferences.disabled_skills",+ "control-openhands browser reload",+ "control-openhands browser fill 'testid=skills-search-input' qa-toggle",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250 --observe 'testid=skill-toggle-qa-toggle' --observe-ms 8000",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands api GET /api/settings --pick misc_settings.app_preferences.disabled_skills",+ "control-openhands browser reload",+ "control-openhands browser fill 'testid=skills-search-input' qa-toggle",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands browser screenshot --feature F18.toggle --name 15694-disabled-persists",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands doctor"+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F18.toggle/15694-disabled-persists.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "The disabled skill stayed off in two eight-second samples and reloads; the server retained its disabled name. The report is intermittent, so longer or different timing remains unverified.",+ "replay_setup": {+ "working_directory": "<checkout>",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations."+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/15694#issuecomment-6121789433"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15701,+ "title": "[Bug]: Creating Conversation toast stuck visible",+ "url": "https://github.com/OpenHands/OpenHands/issues/15701",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.launch-in-workspace"+ ],+ "basis": "known-issue",+ "reason": "The original stuck creation request/outcome sequence; the report is intermittent and gives no deterministic trigger.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15778,+ "title": "LLM config accepts invalid data.",+ "url": "https://github.com/OpenHands/OpenHands/issues/15778",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.create-validation"+ ],+ "basis": "live-reproduced",+ "reason": "Fast refused-loopback validation rejected the profile, but two validations with five retries reached the 30-second browser timeout and then saved with 201.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Fast refused-loopback validation rejected the profile, but two validations with five retries reached the 30-second browser timeout and then saved with 201.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F10.create-validation/15778-invalid-rejected.png",+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved.png",+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved-repeat.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15782,+ "title": "[Bug]: Prompt disappears when a conversation is stopped",+ "url": "https://github.com/OpenHands/OpenHands/issues/15782",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.draft-persistence"+ ],+ "basis": "known-issue",+ "reason": "Clarify whether prompt means unsent composer text or the already-sent history message, and capture that state before sidebar Stop.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15813,+ "title": "[Bug]: Condenser Max Number of Events indicator can go under 0",+ "url": "https://github.com/OpenHands/OpenHands/issues/15813",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F15",+ "feature_ids": [+ "F15.save-errors"+ ],+ "basis": "live-reproduced",+ "reason": "Max Size accepted -5 and enabled Save twice. The server rejected the save and kept 240; the client input floor is missing.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "backend": "localAgentServer1.54.0",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "passes": 2,+ "viewport": "1440x1000",+ "steps": [+ "Max Size accepted -5 and enabled Save twice. The server rejected the save and kept 240; the client input floor is missing.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F15.save-errors/15813-negative-enabled.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15811,+ "title": "[Bug]: Verification Settings Base/Advanced are the same",+ "url": "https://github.com/OpenHands/OpenHands/issues/15811",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F15",+ "feature_ids": [+ "F15.verification-page",+ "F15.verification-critic-fields"+ ],+ "basis": "live-reproduced",+ "reason": "Default Basic and Advanced both show two switches. Enabling Confirmation Mode reveals Security Analyzer only on Advanced. The report asks whether this is intentional; mapped as conditional-tier clarification.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "backend": "localAgentServer1.54.0",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "passes": 2,+ "viewport": "1440x1000",+ "steps": [+ "Default Basic and Advanced both show two switches. Enabling Confirmation Mode reveals Security Analyzer only on Advanced. The report asks whether this is intentional; mapped as conditional-tier clarification.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F15.verification-page/15811-basic.aria.txt",+ "backfill-18309-settings-worker/evidence/F15.verification-page/15811-advanced.aria.txt",+ "backfill-18309-settings-worker/evidence/F15.verification-critic-fields/15811-conditional-advanced.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15822,+ "title": "Bug: Cloud-proxy calls continue after switching to local backend",+ "url": "https://github.com/OpenHands/OpenHands/issues/15822",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.switch-backend",+ "F25.health-status",+ "F25.cloud-org-rows"+ ],+ "basis": "known-issue",+ "reason": "A real registered authenticated Cloud backend/account/organization is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15833,+ "title": "[Bug]: Missing text chat when Create Automation button is clicked in modal",+ "url": "https://github.com/OpenHands/OpenHands/issues/15833",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "not-reproduced",+ "family": "F21",+ "feature_ids": [+ "F21.add-menu",+ "F22.agent-assisted-setup"+ ],+ "reason": "Add Automation → Create Automation opens /automations/setup with a New Automation draft editor, prompt, schedule and assistance input. The old modal-to-empty-conversation branch is replaced on this target, so its missing prefilled chat text does not reproduce.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Fresh doctor/onboarding; browser goto /automations.",+ "browser click testid=automations-add-automation; browser click testid=automations-add-automation-create --expect-url /automations/setup.",+ "Take screenshot and full browser snapshot (setup has no main landmark).",+ "Observed New Automation draft editor with Name, Prompt, Schedule, model/profile and assistance input; the old Create Automation modal-to-empty-chat branch has been replaced."+ ],+ "artifacts": [+ "evidence/F22.agent-assisted-setup/create-flow-15833.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/15833#issuecomment-6121793408"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15946,+ "title": "Show conversations hidden by active/stale agent-server leases",+ "url": "https://github.com/OpenHands/OpenHands/issues/15946",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "basis": "known-issue",+ "reason": "A copied lease-bearing persisted conversation and second owner/expired-lease state; never alter another live run's state.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5718,+ "title": "[Bug]: Plugin or skill loading error should not prevent conversation from starting",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5718",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F19",+ "feature_ids": [+ "F19.enabled-autoload"+ ],+ "basis": "live-reproduced",+ "reason": "Two malformed-manifest first sends showed Invalid JSON and Sending, persisted no events and stayed idle. A malformed skill under a valid manifest instead initialized successfully.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Two malformed-manifest first sends showed Invalid JSON and Sending, persisted no events and stayed idle. A malformed skill under a valid manifest instead initialized successfully.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F19.enabled-autoload/5718-broken-manifest-first-send.png",+ "backfill-18309-settings-worker/evidence/F19.enabled-autoload/5718-broken-manifest-repeat.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5711,+ "title": "[Bug]: TypeScript RemoteState can retain stale snapshots and overwrite newer updates",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5711",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Canvas does not use TypeScript RemoteState; it owns event/state hydration through its frontend stores."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5652,+ "title": "[Bug]: RemoteState confirmation_policy/security_analyzer/agent fail or return None on the second read",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5652",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Python RemoteConversation cached getter behavior; Canvas consumes typed HTTP/WebSocket payloads rather than Python RemoteState."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5651,+ "title": "[Bug]: ConversationEventStream on Node 22 WebSocket stops after one refused reconnect instead of backing off",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5651",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Node 22 ConversationEventStream reconnect path; Canvas uses browser sockets and requires Node >=24 for its launcher."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5650,+ "title": "[Bug]: TS RemoteWorkspace.downloadAsBlob/fileDownload return 11 mangled bytes for a 5-byte binary file",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5650",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "RemoteWorkspace.downloadAsBlob/fileDownload are not Canvas download consumers; Canvas uses binary-safe FileClient.downloadFile."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5649,+ "title": "[Bug]: TS ConversationManager.createConversation ignores maxIterations/stuckDetection (server gets 500/true)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5649",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Canvas builds its create payload and calls ConversationClient directly, not TypeScript ConversationManager.createConversation."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5648,+ "title": "[Bug]: /app-backends WebSocket bridge negotiates no subprotocol (expected the backend's choice, qa.v1)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5648",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F20",+ "feature_ids": [+ "F20.app-backend-view"+ ],+ "basis": "known-issue",+ "reason": "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas."+ ],+ "artifacts": [+ "evidence/F20.app-backend-view/bridge-unavailable.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5647,+ "title": "[Bug]: Continued /v1/chat/completions turn reports conversation-total usage (13/7/20), not its own (3/2/5)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5647",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "OpenAI-compatible gateway per-turn usage contract; Canvas does not consume /v1/chat/completions gateway usage."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5646,+ "title": "[Bug]: Canvas App backend stays 'unhealthy' after one slow health probe instead of returning to 'ready'",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5646",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F20",+ "feature_ids": [+ "F20.app-backend-view"+ ],+ "basis": "known-issue",+ "reason": "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas."+ ],+ "artifacts": [+ "evidence/F20.app-backend-view/bridge-unavailable.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5645,+ "title": "[Bug]: Skill/Canvas install with an unknown ref returns 200 when the git source is cached (expected 400)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5645",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F19",+ "feature_ids": [+ "F19.add-plugin"+ ],+ "basis": "live-reproduced",+ "reason": "Two Add attempts with an unknown ref on a warm local Git cache succeeded and installed cached version 1.0.0 instead of rejecting the missing ref.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Two Add attempts with an unknown ref on a warm local Git cache succeeded and installed cached version 1.0.0 instead of rejecting the missing ref.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F19.add-plugin/5645-unknown-ref.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5644,+ "title": "[Bug]: /api/profiles/{name}/validate ignores provider_connection_id: linked draft valid:false, expected true",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5644",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "known-issue",+ "reason": "Canvas's LlmSettingsLocalView.handleSave explicitly skips ProfilesService.validateProfile when connectionId is set. The linked-draft validation request in this SDK report is not emitted by the mapped Canvas profile-save path.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5643,+ "title": "[Bug]: POST /api/settings/mcp/{key} with {} returns 201 and new conversations lose all MCP tools; expected 422",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5643",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F17",+ "feature_ids": [+ "F17.custom-add",+ "F17.agent-uses-server"+ ],+ "basis": "live-reproduced",+ "reason": "An empty MCP body returned 201 twice and persisted enabled:true; real UI-started initialization lost the healthy echo tool. The invalid entry had no UI card.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "An empty MCP body returned 201 twice and persisted enabled:true; real UI-started initialization lost the healthy echo tool. The invalid entry had no UI card.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F17.runtime-tools/5643-mixed-empty-init.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5642,+ "title": "[Bug]: git changes returns 400 for a modified file with a space in its name and mangles non-ASCII names",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5642",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "basis": "live-reproduced",+ "reason": "Tracked my notes.md modification makes scoped git/changes return400 and hides changes.",+ "steps": [+ "In a disposable Git fixture, create and commit tracked my notes.md and café.txt; modify my notes.md.",+ "Open empty Local workspace conversation > right-panel-toggle > conversation-tab-commits.",+ "browser network --filter /git/ => conversation-scoped git/changes status400.",+ "Settled screenshot shows Uncommitted changes absent; only space-name case driven."+ ],+ "evidence": [+ "evidence/F08.uncommitted-changes/sdk-5642-settled.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5641,+ "title": "[Bug]: GET /api/git/changes returns [] for a non-repo directory holding a repository, not its changes",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5641",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "basis": "live-reproduced",+ "reason": "Plain parent workspace shows No commits yet while its child repo has modified/untracked files; removed only run-owned auto-created empty .git to arrange the reported layout.",+ "steps": [+ "Arrange qa-nested/proj Git repo with committed README.md, modify README.md and add notes.txt.",+ "workspace open <run>/workspace/qa-nested; create empty conversation through sidebar workspace launch.",+ "Server creates empty qa-nested/.git; remove only that disposable empty .git to restore the reported non-repo parent layout.",+ "browser reload; open right panel on Commits; browser wait-text \"No commits yet\".",+ "browser network /git/ => scoped changes200; real child repo remains dirty."+ ],+ "evidence": [+ "evidence/F08.uncommitted-changes/sdk-5641-settled.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5640,+ "title": "[Bug]: download-trajectory returns 404 after POST /api/init sets conversations_path (expected 200 zip)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5640",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.download-zip"+ ],+ "basis": "known-issue",+ "reason": "A deferred-init warm-pool Agent Server initialized with a non-default conversations_path; this launcher starts an already-initialized server.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5639,+ "title": "[Bug]: LLM switched in before the first run is not added to ConversationStats; its tokens and cost are lost",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5639",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.llm-profile-picker"+ ],+ "basis": "known-issue",+ "reason": "A working model, a profile switch before the first run, and nonzero usage/cost after that turn.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5638,+ "title": "[Bug]: execute_bash_command returns only the last chunk of output over 1 MiB; null exit_code past 100 chunks",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5638",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-tree"+ ],+ "basis": "live-reproduced",+ "reason": "A real1800-file listing over1MiB drops README.md and creates a partial-path bb folder before the2000-file cap.",+ "steps": [+ "In a disposable repository, arrange1800 small files beneath three220-character directory segments; total file count remains below2000 and sorted listing exceeds1MiB.",+ "Open its empty Local conversation > Files > files-tab-refresh.",+ "browser count role=button[name=\"README.md\"][exact] =>0 though real README.md remains.",+ "Files snapshot shows a spurious partial-path bb folder alongside surviving tail entries; retained screenshot records tree."+ ],+ "evidence": [+ "evidence/F08.files-tree/sdk-5638-settled.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "run_id": "backfill-18309-surface-worker"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5637,+ "title": "[Bug]: Canvas backend outlives a SIGKILLed agent-server; restart reports it stopped and starts a second one",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5637",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F20",+ "feature_ids": [+ "F20.app-backend-view"+ ],+ "basis": "known-issue",+ "reason": "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas."+ ],+ "artifacts": [+ "evidence/F20.app-backend-view/bridge-unavailable.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5636,+ "title": "[Bug]: switch_llm returns 200 but keeps the old LLM when the posted usage_id is already registered",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5636",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "Canvas direct profile switches assign a new profile:<name>:<uuid> usage_id; the raw switch_llm default/already-registered usage_id collision is not its request shape."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5635,+ "title": "[Bug]: Deleting or sandbox-pausing a conversation leaves its open WebSockets silent instead of closing them",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5635",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.reconnect"+ ],+ "basis": "live-reproduced",+ "reason": "prepare-for-sandbox-pause returned204 but the original communicating session socket stayed open for over10s. A newly opened Canvas tab received a new run:false message; the old tab did not. DELETE returned200 and GET404, but both held sockets stayed open over10s. Canvas later redirected through metadata polling; no claim of permanently stale navigation is made.",+ "steps": [+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser goto /conversations/<id>; POST a unique run:false user marker and wait for it; observe the communicating socket connectionId through browser events.",+ "control-openhands --run \"$OH_VERIFY_RUN\" api POST /api/conversations/prepare-for-sandbox-pause --write --data '{}'; observe socket metadata for at least10s without navigating/reloading the original tab.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser tab new /conversations/<id>; wait for original marker. POST QA_PAUSE_5635 as a run:false user event. New tab receives it; tab0 read-only eval document.body.innerText.includes(\"QA_PAUSE_5635\") was false.",+ "control-openhands --run \"$OH_VERIFY_RUN\" api DELETE /api/conversations/<id> --write; observe both original socket connection IDs for at least10s; GET /api/conversations/<id> returns404. Record later UI polling redirect separately.",+ "Save redacted browser events, both tab screenshots, and close/release only owned tabs."+ ],+ "actual": "prepare-for-sandbox-pause returned204 but the original communicating session socket stayed open for over10s. A newly opened Canvas tab received a new run:false message; the old tab did not. DELETE returned200 and GET404, but both held sockets stayed open over10s. Canvas later redirected through metadata polling; no claim of permanently stale navigation is made.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-delete-sockets-5635.json",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-new-tab-5635.png",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-stale-tab-5635.png",+ "backfill-18309-surface-rechecks/evidence/F06.reconnect/deleted-stale-tab-5635.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5633,+ "title": "[Bug]: Failed run blocks the event loop while logging a rich traceback: /alive takes ~5 s, not ~100 ms",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5633",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.remote-backend"+ ],+ "basis": "live-reproduced",+ "reason": "Temporarily replaced only the owned Agent Server with the same pinned interpreter and state under genuine default Rich logging (LOG_JSON/CI/LOG_RICH_TRACEBACKS unset). Canvas Send uses tools=[] and a dummy LLM pointing at closed 127.0.0.1:9; 60 concurrent direct /alive probes had maximum latency 2.091 s while the conversation ended error. No provider reachable. Restored normal launcher and doctor afterward.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Default Rich logging responsiveness (`F26.remote-backend`).** Temporarily replace only an owned Agent Server with its same pinned interpreter/state on the same port; unset `LOG_JSON`, `CI`, `GITHUB_ACTIONS` and `LOG_RICH_TRACEBACKS`, and retain its PID for cleanup. Create a bare-tool conversation whose dummy LLM points at closed `http://127.0.0.1:9/v1` with zero retries. Start 60 direct `/alive` probes concurrently with a genuine Canvas **Send message**; inspect the rendered connection error and final conversation status. Expected: every probe stays below 1.5 s during failure. Known failure, reproduced 2026-10-12 on Agent Server 1.54.0: the conversation ends `error` and one health probe stalls 2.091 s ([OpenHands/software-agent-sdk#5633](https://github.com/OpenHands/software-agent-sdk/issues/5633)). This is a finite connection-error case, separate from the unprovided cyclic websocket fixture in SDK #5373. Stop that owned replacement, restore the normal launcher and doctor."+ ],+ "artifacts": [+ "evidence/F26.remote-backend/rich-failure-latency-5633.json",+ "evidence/F26.remote-backend/rich-failure-5633.png"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5632,+ "title": "[Bug]: Waiting on a locked profile or connection store freezes the whole server; /alive stalls up to 30 s",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5632",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F11",+ "feature_ids": [+ "F11.list"+ ],+ "basis": "live-reproduced",+ "reason": "The own provider-store lock delayed unrelated /alive responses by 25.416 and 28.060 seconds; the second list request returned 503 after 30.052 seconds. Locks were released and doctor passed.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "The own provider-store lock delayed unrelated /alive responses by 25.416 and 28.060 seconds; the second list request returned 503 after 30.052 seconds. Locks were released and doctor passed.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F11.list/5632-lock-blocks-ui.png",+ "backfill-18309-settings-worker/evidence/F11.list/store-lock-diagnostic.json"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5631,+ "title": "[Bug]: events/search and /count with kind=MessageEvent return nothing; only the full class path matches",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5631",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "SDK event-search short-kind alias contract; current Canvas event history requests do not send that short kind filter."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5630,+ "title": "[Bug]: Sub-agent file naming a missing skill makes the first message return 500 instead of being skipped",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5630",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.launch-in-workspace",+ "F03.create-error-toast"+ ],+ "basis": "live-reproduced",+ "reason": "Owned workspace has .agents/agents/qa-bad-skill.md naming nonexistent qa-f27-missing-skill. With only a closed-loopback dummy LLM profile and autotitling disabled, Home launch in that folder returns the missing-skill error and remains on Home. Two genuine message POSTs with run:false and a create with initial_message run:false each returned HTTP 500 before inference.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Missing sub-agent skill (`F03.launch-in-workspace`, `F03.create-error-toast`).** In the run-owned `<run>/workspace/qa-bad-skill/.agents/agents/qa-bad-skill.md`, arrange YAML frontmatter with `name: qa-bad-skill`, `description: Bad skill` and `skills: qa-f27-missing-skill`, followed by body `Body`; do not install that missing skill. Record the original active LLM profile and autotitle setting. Arrange a dummy `qa-offline` profile through `control-openhands api POST /api/profiles/qa-offline --write --data '@<dummy-profile-json>'`: `llm` contains model `openai/qa-offline`, a dummy API key, base URL `http://127.0.0.1:9`, `num_retries: 0`, `timeout: 2` and `usage_id: qa-offline`, with `include_secrets: true`; activate it through `api POST /api/profiles/qa-offline/activate --write --data '{}'`. Set `conversation_settings_diff.autotitle` to `false` through `api PATCH /api/settings --write --data '@<settings-json>'`, so no provider call is possible. From Home open `testid=open-workspace-button`, `testid=workspace-dropdown` and `testid=add-workspaces-button`; use `testid=folder-browser-up` twice, `testid=folder-browser-entry-workspace`, `testid=folder-browser-entry-qa-bad-skill`, `testid=folder-browser-use` and `testid=workspace-launch-button` to confirm that folder. Fill `testid=chat-input` with `QA missing skill creation check`, then `control-openhands browser click 'testid=home-chat-launcher >> testid=submit-button' --observe 'role=status' --observe-ms 3000`. Expected: an invalid sub-agent is skipped with a warning or reported by an actionable 4xx; it does not turn conversation creation into a 500. Known failure (reproduced 2026-10-12, [OpenHands/software-agent-sdk#5630](https://github.com/OpenHands/software-agent-sdk/issues/5630)): `Creating conversation` is replaced by `Skill 'qa-f27-missing-skill' not found but was given to agent 'qa-bad-skill'.`; the loading state ends, the prompt remains and the URL stays `/`. Genuine message requests with `run: false` and creation with an initial message also return 500 before inference. Verified in `backfill-18309-integrations-worker` on `fa1f724ab5737a17453ad6081129eb91c5a423c6`, SDK 1.54.0; evidence `F03.create-error-toast/missing-skill-home-5630-2.png`. Remove only this run-owned agent file and test conversations afterwards, restore the original autotitle setting and active profile, then delete the dummy profile."+ ],+ "artifacts": [+ "evidence/F03.create-error-toast/missing-skill-home-5630-2.png",+ "evidence/F03.create-error-toast/bad-skill-5630.aria.txt"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5629,+ "title": "[Bug]: A second server's first load of a shared conversation reverts meta.json to its stale startup copy",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5629",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.rename-inline"+ ],+ "basis": "live-reproduced",+ "reason": "B was started with shared storage before A renamed an idle conversation from BEFORE_B to AFTER_B. A PATCH and disk confirmed AFTER_B. After own A graceful stop, native B events/count returned200 and rewrote disk title to its BEFORE_B startup copy. Real Add Backend→B card showed BEFORE_B in Canvas. The native first-load request is explicit; an earlier sandbox-pause variant did not reach a valid first load and is not the claimed reproduction.",+ "steps": [+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "On A PATCH title QA_SHARED_BEFORE_B_5629_REPEAT. Start pinned SDK Python: python .agents/skills/verify-openhands/scripts/fixtures/shared-storage-server.py --run \"$OH_VERIFY_RUN\" --seconds 300 (background; no editor/frontend/browser; bounded owned PGID cleanup). It shares only this run's conversations/cipher/session key and records its loopback port in private/shared-server-5629/server.json.",+ "After B listens, A api PATCH /api/conversations/<id> --write --data '{\"title\":\"QA_SHARED_AFTER_B_5629_REPEAT\"}'; read only title/updated_at from shared meta.json and confirm AFTER_B.",+ "control-openhands --run \"$OH_VERIFY_RUN\" service stop agent-server; invoke pinned SDK Python shared-storage-server.py --run \"$OH_VERIFY_RUN\" --load-id <id> (native B events/count, observed200/count0). Immediately read shared meta.json: BEFORE_B was restored.",+ "control-openhands --run \"$OH_VERIFY_RUN\" restart; control-openhands --run \"$OH_VERIFY_RUN\" browser goto /?backend=default-local; control-openhands --run \"$OH_VERIFY_RUN\" doctor (A is restarted only to serve Canvas).",+ "Through real Add Backend select Agent-server, name QA_Shared_Repeat_5629, host=http://127.0.0.1:<B-port>, key via browser fill --value-file \"$OH_VERIFY_RUN\"/private/session-key, Connect, onboard --skip if needed, then click its BEFORE_B card. conversation-name-title displays BEFORE_B.",+ "Stop only the helper operator PID recorded in its own metadata (SIGTERM); verify its PGID/port gone. Return Canvas to default-local and eventually stop the owned primary."+ ],+ "actual": "B was started with shared storage before A renamed an idle conversation from BEFORE_B to AFTER_B. A PATCH and disk confirmed AFTER_B. After own A graceful stop, native B events/count returned200 and rewrote disk title to its BEFORE_B startup copy. Real Add Backend→B card showed BEFORE_B in Canvas. The native first-load request is explicit; an earlier sandbox-pause variant did not reach a valid first load and is not the claimed reproduction.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-repeat-before-b-load-5629.json",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-repeat-after-b-load-5629.json",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/shared-first-load-reverted-5629.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5628,+ "title": "[Bug]: Resumed conversation keeps a since-disabled plugin's skills and commands; only its hooks are dropped",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5628",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F19",+ "feature_ids": [+ "F19.enabled-autoload"+ ],+ "basis": "live-reproduced",+ "reason": "Two actual Local conversations initialized an enabled installed plugin, then resumed after real UI disable plus server restart. Each resumed command still activated qa-resume-plugin:qa-hello and include_skills retained both plugin skills, while the initial hook count stayed one. A fresh chat begun while disabled had no command activation or hook. All messages reached only refused loopback model initialization; command activation and retained state are pre-model observables, not successful command instructions or inference.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Arrange the run-owned qa-resume-plugin directory with the exact .plugin/plugin.json, commands/qa-hello.md, skills/qa-plugin-skill/SKILL.md and hooks/hooks.json specs captured in live.fixture_specs; hook command is only echo QA_HOOK.",+ "control-openhands api POST /api/plugins/install --write --data @backfill-18309-settings-rechecks/<private-fixture>",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events 47d3727b-9349-4579-b938-c1f9705f80b5 --kinds HookExecutionEvent --from-start --last 4",+ "control-openhands api GET '/api/conversations/47d3727b-9349-4579-b938-c1f9705f80b5?include_skills=true' --pick agent.agent_context.skills",+ "control-openhands browser goto /plugins",+ "control-openhands browser testids --filter qa-resume",+ "control-openhands browser click 'testid=plugin-toggle-qa-resume-plugin' --hover-first 250",+ "control-openhands browser attr 'testid=plugin-toggle-qa-resume-plugin' aria-checked",+ "control-openhands api GET /api/plugins/installed",+ "control-openhands browser goto /conversations/47d3727b-9349-4579-b938-c1f9705f80b5",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands browser reload",+ "control-openhands api GET /api/plugins/installed",+ "control-openhands conversation send 47d3727b-9349-4579-b938-c1f9705f80b5 --prompt '/qa-resume-plugin:qa-hello'",+ "control-openhands conversation events 47d3727b-9349-4579-b938-c1f9705f80b5 --kinds MessageEvent,HookExecutionEvent --last 5",+ "control-openhands api GET '/api/conversations/47d3727b-9349-4579-b938-c1f9705f80b5?include_skills=true' --pick agent.agent_context.skills > 'backfill-18309-settings-rechecks/<private-fixture>'",+ "control-openhands conversation start --prompt '/qa-resume-plugin:qa-hello' --timeout 60",+ "control-openhands api PATCH /api/plugins/installed/qa-resume-plugin --write --data '{\"enabled\":true}'",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands browser goto /plugins",+ "control-openhands browser click 'testid=plugin-toggle-qa-resume-plugin' --hover-first 250",+ "control-openhands api GET /api/plugins/installed",+ "control-openhands conversation events 6c606d12-4fc7-42bd-9dee-9792e0277e38 --kinds HookExecutionEvent --from-start --last 4",+ "control-openhands browser goto /conversations/6c606d12-4fc7-42bd-9dee-9792e0277e38",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands browser reload",+ "control-openhands conversation send 6c606d12-4fc7-42bd-9dee-9792e0277e38 --prompt '/qa-resume-plugin:qa-hello'",+ "control-openhands conversation events 6c606d12-4fc7-42bd-9dee-9792e0277e38 --kinds MessageEvent,HookExecutionEvent --last 5",+ "control-openhands conversation events cc4cb574-1e28-44e4-883c-f36f73161242 --kinds MessageEvent,HookExecutionEvent --last 5",+ "control-openhands browser screenshot --feature F19.enabled-autoload --name 5628-disabled-resume-activation",+ "control-openhands api GET '/api/conversations/6c606d12-4fc7-42bd-9dee-9792e0277e38?include_skills=true' --pick agent.agent_context.skills > 'backfill-18309-settings-rechecks/<private-fixture>'"+ ],+ "evidence": [+ "backfill-18309-settings-rechecks/evidence/F19.enabled-autoload/resume-plugin-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F19.enabled-autoload/5628-disabled-resume-activation.png"+ ],+ "fixture_specs": {+ "plugin": "qa-resume-plugin",+ "manifest": {+ "name": "qa-resume-plugin",+ "version": "0.1.0",+ "description": "Owned bug-backfill resume fixture."+ },+ "command": "commands/qa-hello.md: description frontmatter + Do not run tools. Reply with the single word QA_COMMAND.",+ "skill": "skills/qa-plugin-skill/SKILL.md: name/description frontmatter + Do not run tools. Reply with the single word QA_SKILL.",+ "hook": {+ "hooks": {+ "UserPromptSubmit": [+ {+ "hooks": [+ {+ "type": "command",+ "command": "echo QA_HOOK"+ }+ ]+ }+ ]+ }+ }+ },+ "observation": "Persisted disabled plugin skills/command activation survived both resumes, while hooks were dropped; the fresh disabled conversation was the control.",+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ },+ "run_id": "backfill-18309-settings-rechecks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5627,+ "title": "[Bug]: /goal loop active at a graceful restart stays `running` afterwards instead of `interrupted`",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5627",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.slash-goal"+ ],+ "basis": "known-issue",+ "reason": "A working model goal loop active during graceful backend restart, then goal-status and resume readback.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5626,+ "title": "[Bug]: Pause during a model call not saved: GET says running (expected paused); after SIGKILL it is error",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5626",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "family": "F06",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "basis": "known-issue",+ "reason": "Actual Canvas Stop displayed Stopped; GET and base_state.json both immediately reported paused while a dummy loopback model request was held. The exact launcher-owned Agent Server PID was SIGKILLed, then the own run restarted; GET, disk and reloaded UI retained paused. The accept-only fixture received one connection, generated no response, and bounded-exited cleanly.",+ "steps": [+ "Recorded actions and replay guidance: export OH_VERIFY_RUN to the full directory returned by a new isolated launch; replace conversation IDs and fixture paths with those returned by that run. The recorded run and measurements below are unchanged. JSON bodies shown after --data are shell-quoted; fixture JSON must be saved before an @file command.",+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "python .agents/skills/verify-openhands/scripts/fixtures/accept-only-endpoint.py --output-dir \"$OH_VERIFY_RUN\"/private/accept-only-5626 --seconds 180 (background; reads no request bytes)",+ "Read private port.json; create dummy hold profile model=openai/qa-hold, dummy key, base_url=http://127.0.0.1:<port>/v1, num_retries=0, timeout 600 seconds; activate via own api profile endpoints. Create idle tools[] conversation using that profile.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser goto /conversations/<id>; control-openhands --run \"$OH_VERIFY_RUN\" browser fill testid=chat-input \"QA bounded pause persistence check\"; control-openhands --run \"$OH_VERIFY_RUN\" browser click \"testid=interactive-chat-box >> testid=submit-button\"",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser wait testid=stop-button; control-openhands --run \"$OH_VERIFY_RUN\" api GET /api/conversations/<id> --pick execution_status (running). Confirm accepted_connections=1 in the bounded endpoint metadata.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser click testid=stop-button; control-openhands --run \"$OH_VERIFY_RUN\" browser wait-text Stopped; control-openhands --run \"$OH_VERIFY_RUN\" api GET /api/conversations/<id> --pick execution_status (paused). Read only execution_status from \"$OH_VERIFY_RUN\"/private/state/dev_conversations/<id-hex>/base_state.json (paused).",+ "Find the own run Agent Server listener PID, verify ancestry includes its run.json launcherPid, and SIGKILL only that PID. Retain PID/ancestry proof; never kill by pattern.",+ "control-openhands --run \"$OH_VERIFY_RUN\" restart; control-openhands --run \"$OH_VERIFY_RUN\" doctor; control-openhands --run \"$OH_VERIFY_RUN\" api GET /api/conversations/<id> --pick execution_status (paused); control-openhands --run \"$OH_VERIFY_RUN\" browser reload (Stopped). Do not press Play.",+ "The TCP fixture closes all accepted sockets at its180-second bound; verify its closed metadata and port before final own-run stop."+ ],+ "actual": "Actual Canvas Stop displayed Stopped; GET and base_state.json both immediately reported paused while a dummy loopback model request was held. The exact launcher-owned Agent Server PID was SIGKILLed, then the own run restarted; GET, disk and reloaded UI retained paused. The accept-only fixture received one connection, generated no response, and bounded-exited cleanly.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/bounded-inflight-5626.png",+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/pause-durability-5626.json",+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/owned-crash-5626.json",+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/pause-after-crash-5626.json",+ "backfill-18309-surface-rechecks/evidence/F06.stop-resume/paused-after-crash-5626.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "run_id": "backfill-18309-surface-rechecks",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/5626#issuecomment-6122465657"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5624,+ "title": "[Bug]: Provider connection POST/PATCH accepts blank or redacted api_key values",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5624",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F11",+ "feature_ids": [+ "F11.create-validation",+ "F11.edit"+ ],+ "basis": "live-reproduced",+ "reason": "Two real Add forms accepted the redacted placeholder and persisted keyless connections. Typing it in Edit changed key-set to true. Keyboard Save was used; whitespace-only and provider-auth branches were not claimed.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Two real Add forms accepted the redacted placeholder and persisted keyless connections. Typing it in Edit changed key-set to true. Keyboard Save was used; whitespace-only and provider-auth branches were not claimed.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F11.create-validation/5624-sentinel-created.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5622,+ "title": "[Bug]: After a cipher key change, settings and secret writes answer 200 and null stored keys, not 409",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5622",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F14",+ "feature_ids": [+ "F14.create",+ "F14.edit",+ "F14.delete"+ ],+ "basis": "live-reproduced",+ "reason": "With an approved replacement-cipher/old-encrypted-state fixture, secret creation rewrote an old entry and its dummy getter returned 404. Two unrelated settings saves cleared the old LLM key. Original cipher and stores were restored immediately; restart and doctor passed.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "With an approved replacement-cipher/old-encrypted-state fixture, secret creation rewrote an old entry and its dummy getter returned 404. Two unrelated settings saves cleared the old LLM key. Original cipher and stores were restored immediately; restart and doctor passed.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F14.create/5622-before-cipher-change.png",+ "backfill-18309-settings-worker/evidence/F14.create/5622-after-new-secret.png",+ "backfill-18309-settings-worker/evidence/F14.create/5622-settings-save.png",+ "backfill-18309-settings-worker/evidence/F14.create/cipher-mismatch-diagnostic.json"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F14-secrets.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5607,+ "title": "[Bug]: POST /api/mcp/test classifies a remote MCP server's HTTP 401 as error_kind \"unknown\", so Agent Canvas tells a user with a wrong GitHub token \"Connection failed\"",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5607",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F17",+ "feature_ids": [+ "F17.credential-probe"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5590,+ "title": "[Bug]: api_lifespan ignores the Config passed to create_app(), leaking conversations into OH_CONVERSATIONS_PATH",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5590",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Python create_app(Config) lifespan embedding contract; Canvas launches agent-server using its CLI/environment configuration."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5581,+ "title": "[Bug]: Agent-server startup kills live terminal sessions of other conversations sharing the same `TMUX_TMPDIR`",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5581",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "basis": "known-issue",+ "reason": "Two owned Agent Server runtimes sharing a disposable TMUX_TMPDIR and a model key for the mapped Terminal path are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5539,+ "title": "[Bug]: download-trajectory fails or returns a broken zip when the same trajectory is downloaded again or concurrently",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5539",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.download-zip"+ ],+ "basis": "live-reproduced",+ "reason": "A20-second three-client real loopback overlapping receiver made929 attempts:31 completed corrupt200 ZIPs,106HTTP500,269ingress502,133RemoteProtocolError,388 validZIPs and2 intentionally deadline-aborted reads. Both ambiguous Content-Length differences were excluded from the public claim. Actual Canvas Download during overlap produced a ZIP with inspectable entries; that particular UI download is not claimed corrupt. A separate sequential2000-request pass returned validZIPs throughout.",+ "steps": [+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "Use the pinned SDK Python (httpx installed): python .agents/skills/verify-openhands/scripts/fixtures/download-trajectory-overlap.py --base-url <own-base-url> --conversation-id <id> --api-key-file \"$OH_VERIFY_RUN\"/private/session-key --output-dir \"$OH_VERIFY_RUN\"/private/trajectory-overlap --seconds 20 --clients 3 --rounds 5000 --slow-ms 2 (background). It rejects non-loopback origins, caps network/uncompressed bytes/members, checks CRC in bounded chunks, never treats aborted reads as broken, and closes every client.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser goto /conversations/<id>; control-openhands --run \"$OH_VERIFY_RUN\" browser click \"testid=chat-pane-header >> testid=ellipsis-button\"; control-openhands --run \"$OH_VERIFY_RUN\" browser click testid=download-trajectory-button while real receivers overlap.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser downloads --last 3 --inspect; retain the real receiver sanitized summary. ZIP samples and session keys remain private.",+ "Read the public corrected summary:31 completed brokenZIPs supports the race;2 original length differences were ambiguous with2 deadline aborts, so no length/truncation count is asserted."+ ],+ "actual": "A 20-second three-client real loopback overlapping receiver made929 attempts:31 completed corrupt200 ZIPs,106HTTP500,269ingress502,133RemoteProtocolError,388 validZIPs and2 intentionally deadline-aborted reads. Both ambiguous Content-Length differences were excluded from the public claim. Actual Canvas Download during overlap produced a ZIP with inspectable entries; that particular UI download is not claimed corrupt. A separate sequential2000-request pass returned validZIPs throughout.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F07.download-zip/trajectory-overlap-5539-summary.json",+ "backfill-18309-surface-rechecks/evidence/F07.download-zip/trajectory-summary-correction.json",+ "backfill-18309-surface-rechecks/evidence/F07.download-zip/overlapping-download-5539.png"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5533,+ "title": "Tmux can accept commands before slow Bash startup completes",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5533",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "basis": "known-issue",+ "reason": "A model key for real agent-driven first-command observations is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5523,+ "title": "[Bug]: deepseek-v4.1-flash is not detected as vision-capable (supports_vision=False), so its profile is excluded from the vision fallback",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5523",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.image-only-send"+ ],+ "basis": "known-issue",+ "reason": "A working DeepSeek V4.1 Flash vision profile and another fallback profile, with an image-bearing turn.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5519,+ "title": "[Bug]: A scoped profile can't delegate to general-purpose because it lacks task_tracker",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5519",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.openhands-options",+ "F13.active-drives-conversation"+ ],+ "basis": "known-issue",+ "reason": "A complete scoped worker turn needs an approved working model and the affected legacy profile to dispatch a child and invoke TaskTrackingTool. Current Canvas Save added task_tracker despite its unchecked box, and real parent initialization exposed task/task_tracker; this is not a child-turn verdict.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 1,+ "steps": [+ "A complete scoped worker turn needs an approved working model and the affected legacy profile to dispatch a child and invoke TaskTrackingTool. Current Canvas Save added task_tracker despite its unchecked box, and real parent initialization exposed task/task_tracker; this is not a child-turn verdict.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F13.openhands-options/scoped-task-diagnostic.json"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5501,+ "title": "[Bug]: Canvas extension manifest drops pages[].nav_label, so Agent Canvas shows the page title in the rail",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5501",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F20",+ "feature_ids": [+ "F20.rail-label"+ ],+ "basis": "existing-note",+ "reason": "Existing Agent Server nav_label omission note and F20.rail-label recipe preserved.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5498,+ "title": "[Bug]: After deleting the active LLM profile, its provider connection cannot be deleted (\"referenced by the active agent settings\")",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5498",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F11",+ "feature_ids": [+ "F11.delete-stale-reference"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5497,+ "title": "[Bug]: Rotating a provider connection's API key does not reach the active LLM settings; new conversations keep using the old key",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5497",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F11",+ "feature_ids": [+ "F11.rotate"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5496,+ "title": "[Bug]: Concurrent plugin requests corrupt .installed.json, so a just-installed plugin is recorded as source \"local\" and Update/launch then fail",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5496",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F19",+ "feature_ids": [+ "F19.refresh",+ "F19.start-conversation",+ "F19.enabled-autoload"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5495,+ "title": "[Bug]: condenser_kind is exposed as a user-facing settings field with a reStructuredText developer description",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5495",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F15",+ "feature_ids": [+ "F15.condenser-kind"+ ],+ "basis": "existing-note",+ "reason": "Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5492,+ "title": "[Bug]: goal judge treats `\"complete\": \"false\"` (a string) as complete",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5492",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.slash-goal"+ ],+ "basis": "known-issue",+ "reason": "A real goal judge response containing complete as the string false rather than a JSON boolean.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5481,+ "title": "[Bug]:",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5481",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Empty bug template identifies no behavior, reproduction, version or mapped Canvas feature; needs a concrete report before Canvas relevance can be established."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5475,+ "title": "[Bug]: version-bump-prs has never opened the enterprise SDK bump PR (403); land #5339",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5475",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Enterprise SDK dependency-bump CI permissions; no Canvas feature or user flow."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5466,+ "title": "[Bug]: A stale codex-acp / claude-agent-acp on PATH silently replaces the pinned ACP version (Codex ACP fails to start)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5466",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "basis": "live-reproduced",+ "reason": "Two real Canvas Codex ACP starts resolved acp_command=['codex-acp'] and initialized the run-owned published @zed-industries/[email protected] ahead of PATH despite the SDK registry's 1.10.0 pin. Both actual logs reported agent_version=0.16.0 and the mismatch warning still blamed npx fallback. Initialization then stopped at Authentication required, with private HOME and provider routes refused on loopback. This proves stale binary selection/warning before auth, not a usable login or historical model-cache failure.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('backfill-18309-settings-rechecks/<private-fixture>');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "steps": [+ "npm install --prefix 'backfill-18309-settings-rechecks/<private-fixture>' --userconfig 'backfill-18309-settings-rechecks/<private-fixture>' --cache 'backfill-18309-settings-rechecks/<private-fixture>' --no-audit --no-fund @zed-industries/[email protected]",+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=add-agent-profile'",+ "control-openhands browser choose 'testid=agent-type-selector' 'ACP (external subprocess)'",+ "control-openhands browser choose 'testid=agent-preset-selector' Codex",+ "control-openhands browser fill 'testid=agent-profile-name-input' QA_STALE_CODEX",+ "control-openhands browser fill 'testid=settings-acp-secret-OPENAI_API_KEY' --value-file <owned-fixture>",+ "control-openhands browser fill 'testid=settings-acp-secret-OPENAI_BASE_URL' http://127.0.0.1:1/v1",+ "control-openhands browser click 'testid=save-agent-profile-btn'",+ "control-openhands api GET /api/agent-profiles/QA_STALE_CODEX --pick profile.acp_server",+ "control-openhands doctor",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=QA_STALE_CODEX >> testid=agent-profile-menu-trigger'",+ "control-openhands browser testids --filter agent-profile",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\np=Path('backfill-18309-settings-rechecks/<private-fixture>');d=json.loads(p.read_text());d['UV_OFFLINE']='true';d['ALL_PROXY']='http://127.0.0.1:1';d['all_proxy']='http://127.0.0.1:1';p.write_text(json.dumps(d));print({'owned_UV_OFFLINE':True,'owned_ALL_PROXY_refused_loopback':True})\nPY",+ "export HTTP_PROXY=http://127.0.0.1:1 HTTPS_PROXY=http://127.0.0.1:1 http_proxy=http://127.0.0.1:1 https_proxy=http://127.0.0.1:1 NO_PROXY=127.0.0.1,localhost no_proxy=127.0.0.1,localhost\nexport PATH='backfill-18309-settings-rechecks/<private-fixture>':\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"\ncontrol-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands browser reload",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events fab9744c-beb5-4903-945f-9de7cca3ac63 --kinds ConversationErrorEvent --last 2",+ "control-openhands browser screenshot --feature F13.acp-conversation --name 5466-stale-published-adapter",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60"+ ],+ "evidence": [+ "backfill-18309-settings-rechecks/evidence/F13.acp-conversation/stale-adapter-diagnostic.json",+ "backfill-18309-settings-rechecks/evidence/F13.acp-conversation/5466-stale-published-adapter.png"+ ],+ "observation": "Both actual ACP initializations reported the published stale 0.16.0 adapter rather than the pinned 1.10.0, and the warning misattributed selection to npx fallback.",+ "fixture_specs": {+ "package": "@zed-industries/[email protected]",+ "installation": "backfill-18309-settings-rechecks/<private-fixture>",+ "private_HOME": "backfill-18309-settings-rechecks/<private-fixture>",+ "provider_routes": "OPENAI_BASE_URL and OPENAI_API_BASE=refused loopback; HTTP(S)/ALL proxy=refused loopback; NO_PROXY=127.0.0.1,localhost; UV_OFFLINE=true during owned restart"+ },+ "uncovered_branches": [+ "usable authenticated Codex session",+ "historical max-variant model-cache failure",+ "Claude adapter counterpart",+ "positive authenticated pinned-binary control"+ ],+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ },+ "run_id": "backfill-18309-settings-rechecks"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5465,+ "title": "[Bug]: A forked conversation loses its LLM API key and fails its first turn with LLMAuthenticationError",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5465",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.branch-from-here"+ ],+ "basis": "known-issue",+ "reason": "A working cipher-backed model conversation, then fork and a successful turn on the fork.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5447,+ "title": "[Bug]: claude-opus-5-5 requests ending in an assistant (prefill) message fail with Anthropic 400 \"does not support assistant message prefill\"",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5447",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "Anthropic Claude Opus5.5 credentials and a real trailing-assistant-prefill request.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5375,+ "title": "[Bug]: GitLab MCP OAuth fails with localhost redirect URI",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5375",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F17",+ "feature_ids": [+ "F17.install-oauth",+ "F17.agent-uses-server"+ ],+ "basis": "known-issue",+ "reason": "A successful GitLab OAuth login can be followed by an invalid localhost redirect URI on the first ticket-to-code-change MCP action. Reproduction needs the authorized self-hosted GitLab MCP endpoint/account and the ticket-to-code-change plugin/workflow; the reported endpoint is not a public fixture.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5373,+ "title": "agent-server: a logging call can wedge the event loop and amplify memory (rich traceback render)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5373",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.remote-backend"+ ],+ "basis": "known-issue",+ "reason": "A separately hosted default-Rich backend and the original cyclic-exception peer/backlog driver are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5354,+ "title": "[Bug]: Validation errors split a single tool-call response in SDK conversation history",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5354",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.event-groups"+ ],+ "basis": "known-issue",+ "reason": "A live model turn mixing valid, invalid and valid tool calls in one response.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5330,+ "title": "[Bug]: /btw causes prompt-cache invalidation",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5330",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.slash-btw"+ ],+ "basis": "known-issue",+ "reason": "A working caching provider and comparable ordinary-message versus /btw prompt/cache records.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5322,+ "title": "[Bug]: Pager guard from #3660/#3684 doesn't cover TmuxPanePool or WindowsTerminal",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5322",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "basis": "known-issue",+ "reason": "A model key for genuine terminal-tool execution is unavailable; the PowerShell half additionally requires Windows.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5310,+ "title": "[Bug]: Docker conversation runtime cannot resolve configured title LLM profile because per-runtime profile store is empty",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5310",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.docker-conversation-runtime"+ ],+ "basis": "known-issue",+ "reason": "Docker daemon/per-conversation image and real auxiliary-title/ACP credentials are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5307,+ "title": "[Bug]: Docker conversation runtime creates linked worktree across incompatible host/container filesystem namespaces",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5307",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.docker-conversation-runtime"+ ],+ "basis": "known-issue",+ "reason": "Docker daemon/per-conversation image is unavailable; a host/container git fixture is required.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5279,+ "title": "fix(workspace): link attached conversations in automation callbacks",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5279",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.run-linked-conversation"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Cloud automation workspace/callback context and owned stable conversation are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5268,+ "title": "[Bug]: openhands.sdk.conversation.exceptions.ConversationRunError: 'PromptTokensDetailsWrapper' object has no attribute 'cache_creation_tokens'",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5268",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "A real provider response carrying prompt_tokens_details on the affected LiteLLM/SDK combination.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5252,+ "title": "[Testing gap]: No cross-version conversation fixture, so a dependency pin can make older persisted conversations unreadable without CI noticing",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5252",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Cross-version test-fixture gap; no reported user-visible failure to reproduce."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5225,+ "title": "[Bug]: LLMBadRequestError from a poisoned tool_result permanently wedges conversations — no recovery path in Agent.step",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5225",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "A real provider-rejected poisoned image tool_result in a disposable persisted history, followed by retry.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5213,+ "title": "[Bug]: ask_agent 500: `AttributeError: 'PromptTokensDetailsWrapper' object has no attribute 'cache_creation_tokens'` in telemetry `_cache_buckets` with litellm >= 1.95.1",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5213",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.slash-btw"+ ],+ "basis": "known-issue",+ "reason": "A working /btw model response with prompt_tokens_details on an affected LiteLLM version.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5207,+ "title": "OpenRouter: resolve shared connections across runtime paths and preserve structured reasoning",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5207",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F11",+ "feature_ids": [+ "F11.agent-uses-connection",+ "F11.rotate"+ ],+ "basis": "known-issue",+ "reason": "OpenRouter linked profiles can lose connection credentials at raw agent/condenser startup or reuse stale credentials after rotation; opaque reasoning can also be lost on persisted turns. End-to-end checks need an approved OpenRouter connection and a model returning reasoning_details, including the linked-profile switch and reload path; loopback failure does not prove them.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5181,+ "title": "[Bug]: output-cap defaults and explicit reasoning overrides are mishandled",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5181",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.schema-validation"+ ],+ "basis": "known-issue",+ "reason": "The SDK can turn an advertised output limit into a requested default cap and silently drop an explicitly enabled reasoning_effort. The narrowed maintainer scope needs an authenticated OpenRouter/proxy route exposing lower-limit endpoints and capability metadata, or a real endpoint accepting the explicit override; a synthetic empty-metadata response would not establish production behavior.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5168,+ "title": "AttributeError in Telemetry._cache_buckets for providers without prompt caching",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5168",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "A working provider response whose usage has no cache-creation field.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5167,+ "title": "[Bug]: Codex ACP on a headless host: `authenticate(\"chat-gpt\")` does not complete within `ACP_AUTH_TIMEOUT` although Codex is logged in (\"ChatGPT authentication did not complete in time\")",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5167",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "basis": "known-issue",+ "reason": "On a headless Docker host an already usable Codex ChatGPT login can still time out during ACP authenticate. Reproduction needs an approved ChatGPT login in the reported headless/read-only-auth-file deployment and its adapter; a desktop or bogus login cannot establish this conditional trigger.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5152,+ "title": "Unlocked installs resolve browser-use 0.11.13 with mcp 2.x — browser tools silently disappear",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5152",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "not-reproduced",+ "family": "F27",+ "feature_ids": [+ "F27.browser-screenshot"+ ],+ "reason": "The actual unlocked pinned runtime resolves openhands-tools 1.54.0, browser-use 0.11.13, mcp 1.30.0 and fastmcp 3.4.8. Its genuine BrowserToolExecutor constructs successfully using its own interpreter and private HOME. The reported mcp 2.x/list_tools import failure does not occur in this exact dependency set; no model was called.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "From this run's launcher ancestry, identify the actual Agent Server Python executable; never print argv or credentials.",+ "Using that interpreter and the run's private HOME, execute importlib.metadata.version for browser-use, mcp, fastmcp and from openhands.tools.browser_use.impl import BrowserToolExecutor; e=BrowserToolExecutor().",+ "Observed browser-use 0.11.13, mcp 1.30.0, fastmcp 3.4.8 and successful BrowserToolExecutor construction with exit 0. No browser operation or LLM was invoked."+ ],+ "artifacts": [+ "evidence/F27.browser-screenshot/dependency-17383-5152.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/5152#issuecomment-6121793764"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5138,+ "title": "Cloud-proxy (TS client) sends upstream body as object -> 422 from /api/cloud-proxy (Pydantic rejects non-string body)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5138",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Retired TypeScript cloud-proxy body API; current Canvas makes direct Cloud App-API calls and uses typed runtime clients."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5108,+ "title": "[Bug]: Gemini edit corrupts non-UTF-8 content during unrelated edits",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5108",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "An Agent Server configured with the native Gemini edit tool set and a real tool edit of an invalid-UTF8 fixture; Gemini ACP uses a different external executor.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5100,+ "title": "[Bug]: Profile pre-flight validation reports budget-exceeded keys as valid and blocks ~2 min on retries",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5100",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.create-validation"+ ],+ "basis": "live-reproduced",+ "reason": "The slow branch reproduced twice: five-retry loopback validation lasted until the 30-second client abort, then Canvas saved the profile. The separate exhausted-quota 429 branch requires an approved real endpoint.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "The slow branch reproduced twice: five-retry loopback validation lasted until the 30-second client abort, then Canvas saved the profile. The separate exhausted-quota 429 branch requires an approved real endpoint.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved.png",+ "backfill-18309-settings-worker/evidence/F10.create-validation/5100-slow-saved-repeat.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5099,+ "title": "1.48.0: _cache_buckets raises AttributeError on litellm PromptTokensDetailsWrapper",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5099",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "A working model response with LiteLLM PromptTokensDetailsWrapper lacking cache_creation_tokens.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5096,+ "title": "[Bug]: FileEditor replaces symlinks instead of updating the target file",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5096",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A working model invoking canonical file-editor str_replace/insert through a real symlink, then undo.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5094,+ "title": "[Bug]: ACP session/load failure silently starts a fresh session without restoring conversation context",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5094",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.open-by-url"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Codex ACP session whose original session/load fails while persisted history remains.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5092,+ "title": "WAITING_FOR_CONFIRMATION: any user message implicitly approves pending actions on next run()",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5092",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.confirmation-mode"+ ],+ "basis": "known-issue",+ "reason": "A working model proposing a real confirmation-gated marker action, then a revocation user message.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5074,+ "title": "TaskTool child remains running and task mapping is lost after Agent Server process restart",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5074",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A native delegated child still running during an owned server crash and parent recovery.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5072,+ "title": "fix(ts-client): keep Babel override consistent with lockfile",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5072",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "TypeScript Babel override/lockfile consistency for builds; no Canvas runtime feature."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5038,+ "title": "[Bug]: UTF-8 Chinese text misclassified as binary when sampling splits a multibyte character",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5038",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A live canonical file-editor view/edit on the specific UTF8 sample alignment; Canvas Files preview uses a separate NUL sniff, not this validator.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5027,+ "title": "fix: preserve repository identity in PR linked-issue readiness checks",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5027",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "PR linked-issue readiness-check repository identity; repository CI tooling."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4990,+ "title": "[Bug]: /pause reports PAUSED while an ACP agent keeps executing tool calls",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4990",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "basis": "known-issue",+ "reason": "An authenticated Claude Code ACP turn executing separate marker-writing tools while Pause is pressed.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4987,+ "title": "[Bug]: Title generation prompt glues user message to instruction and breaks emoji list formatting",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4987",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "basis": "known-issue",+ "reason": "A working autotitle model and captured title-generation input/output; Canvas only consumes the resulting server title.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4983,+ "title": "[Bug]: system grep fallbacks misinterpret extended regular expressions",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4983",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A runtime without ripgrep and a live agent grep call using extended regex and escaped literal punctuation.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4965,+ "title": "[Bug]: to_chat_dict emits blank text blocks and empty content arrays that strict OpenAI-compatible providers reject",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4965",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "A strict OpenAI-compatible provider and a real history containing blank text or empty content arrays.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4963,+ "title": "[Bug]: Windows PowerShell terminal is permanently poisoned after a terminating error or interrupt",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4963",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "basis": "known-issue",+ "reason": "Native Windows PowerShell terminal runtime is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4934,+ "title": "SDK sends unsupported reasoning_effort to MiniMax M3",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4934",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.schema-validation",+ "F10.default-used"+ ],+ "basis": "known-issue",+ "reason": "Switching to a MiniMax M3 profile can fail because the SDK sends unsupported reasoning_effort. The runtime switch needs approved working credentials for the starting model and MiniMax route; the reported provider rejection cannot be proved with a loopback connection error.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4912,+ "title": "Path-gated test jobs omit openhands-sdk/**, so SDK changes can break downstream tests with every required check green",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4912",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "SDK CI path-gating and downstream test selection; no Canvas runtime feature."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4910,+ "title": "[Bug]: ACP subprocess leaf process orphaned on shutdown on POSIX systems",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4910",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "known-issue",+ "reason": "An owned authenticated ACP provider runtime/command chain is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4901,+ "title": "ACP subprocess leaf process orphaned on shutdown (POSIX) — process-group teardown gap, separate from #4409's Windows fix",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4901",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "known-issue",+ "reason": "An owned authenticated ACP provider runtime/command chain is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4893,+ "title": "agent-server: conversation stuck in execution_status=RUNNING after sandbox is paused/stopped mid-run (unresumable via API)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4893",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.status-menu"+ ],+ "basis": "known-issue",+ "reason": "A disposable sandbox/VM frozen mid-run with a live model and persisted ownership state.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4891,+ "title": "MCP connection failure aborts send_message before the LLM is called",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4891",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "family": "F17",+ "feature_ids": [+ "F17.agent-uses-server"+ ],+ "basis": "live-not-reproduced",+ "reason": "Missing stdio and unreachable HTTP MCP sources both allowed initialization and preserved the healthy sibling tool, then reached the expected loopback model error. No successful reply was claimed.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @<owned-fixture>",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands fixture mcp-server --name qa-settings-echo",+ "control-openhands api POST /api/settings/mcp/qa_settings_echo --write --data @<echo-config.json>",+ "control-openhands browser goto /mcp",+ "control-openhands browser click 'testid=mcp-health-probe-qa_settings_echo' --observe 'testid=mcp-health-label-qa_settings_echo' --observe-ms 6000",+ "control-openhands browser text 'testid=mcp-health-label-qa_settings_echo'",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation events 063a0efe-1766-457f-a27f-11edbb9e4345 --kinds SystemPromptEvent --from-start --last 1",+ "control-openhands api POST /api/settings/mcp/qa_settings_bad --write --data '{\"transport\":\"stdio\",\"command\":\"/qa/no-such-binary\",\"enabled\":true}'",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation events 701d69e9-b476-4170-a165-0742aae18196 --kinds SystemPromptEvent --from-start --last 1",+ "control-openhands conversation status 701d69e9-b476-4170-a165-0742aae18196",+ "control-openhands api DELETE /api/settings/mcp/qa_settings_bad --write",+ "control-openhands api POST /api/settings/mcp/qa_unreachable --write --data '{\"transport\":\"streamable-http\",\"url\":\"http://127.0.0.1:1/mcp\",\"enabled\":true}'",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation events 2653fd2e-6cea-48f9-a15e-a8351fba3cda --kinds SystemPromptEvent --from-start --last 1",+ "control-openhands conversation status 2653fd2e-6cea-48f9-a15e-a8351fba3cda",+ "control-openhands api DELETE /api/settings/mcp/qa_unreachable --write",+ "control-openhands api DELETE /api/settings/mcp/qa_settings_echo --write",+ "control-openhands doctor"+ ],+ "evidence": [],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Missing stdio and unreachable HTTP MCP sources both allowed initialization and preserved the healthy sibling tool, then reached the expected loopback model error. No successful reply was claimed.",+ "replay_setup": {+ "working_directory": "<checkout>",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "<owned-fixture>",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ },+ "echo_config": {+ "file": "<echo-config.json>",+ "body": {+ "transport": "stdio",+ "command": "<Node executable returned by fixture mcp-server>",+ "args": [+ "<server.mjs returned by fixture mcp-server>"+ ],+ "enabled": true+ }+ }+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4891#issuecomment-6121790068"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4887,+ "title": "Centralize release publication dispatches",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4887",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "SDK release publication workflow dispatch; no Canvas feature."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4847,+ "title": "[Bug]: RemoteConversation.fork() ignores title in server response; test mocks wrong server shape",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4847",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Python RemoteConversation.fork() wrapper title parsing; Canvas calls the typed REST fork endpoint directly."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4846,+ "title": "[Bug]: WebSocket clients ignore Node's global WebSocket in ESM builds",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4846",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Node ESM WebSocket constructor selection; Canvas browser selects window.WebSocket."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4811,+ "title": "[Bug]: fork sets title by injecting into state.tags instead of a dedicated field",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4811",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "The report is about standalone SDK fork title injection into ConversationState.tags; it explicitly separates Agent Server StoredConversation title handling used by Canvas."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4810,+ "title": "[Bug]: confirmation_policy, security_analyzer, and secrets mutations not persisted to meta.json",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4810",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "Canvas sets policy/analyzer/secrets at creation or through account settings/resume secrets; it does not call these per-conversation mutation endpoints."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4800,+ "title": "[Bug]: ConversationService: stored fields mutated in memory before save_meta() with no rollback on failure",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4800",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F07",+ "feature_ids": [+ "F07.rename-inline"+ ],+ "basis": "live-reproduced",+ "reason": "After a successful baseline UI rename, chmod0500 on only the owned conversation directory made the next actual UI rename fail with PermissionError/HTTP500. GET nevertheless reported the rejected title while meta.json retained the baseline. Permissions were restored; graceful restart then persisted the rejected in-memory title. Tags/autotitle branches were not tested.",+ "steps": [+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser goto /conversations/<id>; dblclick testid=conversation-name-title, fill conversation-name-input QA_META_BEFORE_4800, press Enter; verify GET title and disk meta title.",+ "Record the original permissions of \"$OH_VERIFY_RUN\"/private/state/dev_conversations/<id-hex>, then chmod 0500 only that owned directory (a read-only meta file alone cannot block an atomic replacement).",+ "Through the same UI rename to QA_META_REFUSED_4800. Observe permission-denied toast/HTTP500; GET title is REFUSED, but read only title/updated_at from meta.json and observe BEFORE.",+ "Restore the directory exact original mode in finally. control-openhands --run \"$OH_VERIFY_RUN\" restart; control-openhands --run \"$OH_VERIFY_RUN\" doctor; GET/reload and disk now show the previously rejected REFUSED title persisted during graceful shutdown.",+ "Retain only sanitized title/time snapshots; no disk-full/global filesystem changes or raw stored secrets."+ ],+ "actual": "After a successful baseline UI rename, chmod0500 on only the owned conversation directory made the next actual UI rename fail with PermissionError/HTTP500. GET nevertheless reported the rejected title while meta.json retained the baseline. Permissions were restored; graceful restart then persisted the rejected in-memory title. Tags/autotitle branches were not tested.",+ "artifacts": [+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-before-4800.json",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-denied-4800.json",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/refused-title-4800.png",+ "backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-after-graceful-restart-4800.json"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "run_id": "backfill-18309-surface-rechecks"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4798,+ "title": "TypeScript client endpoint audit fails on every fork PR: comment POST 403s with a read-only token",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4798",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "TypeScript client endpoint-audit CI comment permissions; no Canvas feature."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4777,+ "title": "[Bug]: ParallelToolExecutor starts a pending tool after cancellation while waiting for a resource lock",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4777",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "basis": "known-issue",+ "reason": "A live batch with a second real tool waiting on a declared-resource lock when Stop is pressed.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4776,+ "title": "[Bug]: MCPToolObservation silently discards structuredContent and result _meta",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4776",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "known-issue",+ "reason": "There is no structuredContent, structured_content, result_meta or resultMeta consumer under current src/. The report demonstrates SDK conversion/persistence metadata loss, without a current mapped Canvas renderer or API consumer for those fields.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4774,+ "title": "[Bug]: MessageEvent.visualize returns blank output for an empty TextContent block",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4774",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Python MessageEvent.visualize() CLI terminal rendering; Canvas renders MessageEvent content in its own React components."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4769,+ "title": "[Bug]: Responses streaming overwrites a yielded completion event with stale wrapper state",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4769",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "A working Responses-mode provider whose stream wrapper yields completion but retains completed_response=None.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4765,+ "title": "[Bug]: redact_text_secrets leaves api_key assignments with spaces around = unredacted",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4765",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "known-issue",+ "reason": "The affected redact_text_secrets helper is used by SDK command and ACP diagnostic logging. This report supplies only a direct Python helper call and no affected Canvas-rendered event, endpoint field or user control; F14 secret CRUD does not call it.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4709,+ "title": "[Bug]: AgentContext.current_datetime persists in settings.json, causing stale CURRENT_DATETIME in prompts",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4709",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "family": "F15",+ "feature_ids": [+ "F15.agent-context-effect"+ ],+ "basis": "live-not-reproduced",+ "reason": "Canvas launches used fresh runtime datetimes despite naturally stale settings and an explicit old-date fixture. The old date was absent from the initialized prompt.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @<owned-fixture>",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands api GET /api/settings --pick agent_settings.agent_context.current_datetime",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "python3 - <<'PY_DATE'\nimport json, os\nfrom pathlib import Path\nfor p in (Path(os.environ['OH_VERIFY_RUN']) / 'private').rglob('base_state.json'):\n d=json.loads(p.read_text())\n print({'conversation':p.parent.name,'runtime_datetime':d.get('agent',{}).get('agent_context',{}).get('current_datetime')})\nPY_DATE",+ "control-openhands api PATCH /api/settings --write --data '{\"agent_settings_diff\":{\"agent_context\":{\"current_datetime\":\"2001-01-01T00:00:00+00:00\"}}}' --pick agent_settings.agent_context.current_datetime",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation events 950e5e83-f56d-4dbb-9aab-b2ccc78f1972 --kinds SystemPromptEvent --grep 2001-01-01 --from-start --last 1",+ "python3 - <<'PY_DATE'\nimport json, os\nfrom pathlib import Path\nfor p in (Path(os.environ['OH_VERIFY_RUN']) / 'private').rglob('base_state.json'):\n d=json.loads(p.read_text())\n print({'conversation':p.parent.name,'runtime_datetime':d.get('agent',{}).get('agent_context',{}).get('current_datetime')})\nPY_DATE",+ "control-openhands api PATCH /api/settings --write --data '{\"agent_settings_diff\":{\"agent_context\":{\"current_datetime\":\"2026-10-12T07:29:16.188868+02:00\"}}}' --pick agent_settings.agent_context.current_datetime",+ "control-openhands doctor"+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F15.agent-context-page/datetime-diagnostic.json"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Canvas launches used fresh runtime datetimes despite naturally stale settings and an explicit old-date fixture. The old date was absent from the initialized prompt.",+ "replay_setup": {+ "working_directory": "<checkout>",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "<owned-fixture>",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ },+ "datetime_restore": "Capture the original settings timestamp with the first GET, then substitute it into the final PATCH. The literal 2026-10-12 value was the observed original in this run."+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4709#issuecomment-6121790419"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4705,+ "title": "[Bug]: Missing GitHub MCP inputSchema causes inconsistent task status between the task list and automation page",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4705",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling",+ "F23.run-logs-modal"+ ],+ "basis": "known-issue",+ "reason": "Affected client/server versions, GitHub tool-definition event fixture and a model key for genuine run/status comparison are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4670,+ "title": "WebSocketCallbackClient.stop() does not close a silent WebSocket and leaves its worker thread running",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4670",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Python WebSocketCallbackClient thread shutdown; not the Canvas browser WebSocket path."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4668,+ "title": "[Bug]: Planning file editor drops inherited observation and diff data",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4668",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A live planner create/edit yielding PlanningFileEditorObservation with old/new contents for the chat diff.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4667,+ "title": "[Bug]: Tom processing history can checkpoint events that were never indexed",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4667",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Tom conversation index/checkpoint implementation; no mapped Canvas feature or supported Canvas agent path identified."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4666,+ "title": "[Bug]: Duplicate browser recording starts can orphan an active flush task",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4666",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "reason": "The report concerns the backend browser-use rrweb recording controller/flush-task lifecycle. Canvas has no backend recording/replay surface or duplicate-start action; control-openhands browser record is a separate verification-browser implementation and is not this SDK consumer.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4665,+ "title": "[Bug]: Skills marketplace returns empty against the current extensions manifest layout",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4665",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "known-issue",+ "reason": "Canvas SkillsService.getSkills uses the bundled @openhands/extensions catalog and the local /api/skills list with load_public:false. It does not call the SDK skills marketplace loader/endpoint whose default.json assumption is reported.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4664,+ "title": "[Bug]: OpenHandsCloudWorkspace resume leaves the active client on stale connection data",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4664",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "OpenHandsCloudWorkspace is a Python workspace object; Canvas uses TypeScript clients and Cloud runtime API transport, not that object or its cached Python connection fields."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4663,+ "title": "[Bug]: Python glob fallback mutates the process-wide working directory",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4663",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "Concurrent real glob/tool calls in distinct workspaces with the Python fallback active.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4661,+ "title": "[Bug]: Process-global subagent registry leaks definitions across conversations",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4661",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "Two native task-enabled conversations with same-name project subagent definitions and a working model for delegation.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4660,+ "title": "[Bug]: RouterLLM bypasses model selection for async completions",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4660",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "basis": "known-issue",+ "reason": "Canvas Model Router uses meta-profiles and buildRouterAtStartSystemSuffix to guide profile selection; it does not instantiate the Python RouterLLM or call its completion/acompletion methods. The reported SDK class path is not its implementation.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4659,+ "title": "[Bug]: Concurrent stdout and stderr chunks can reuse BashOutput order values",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4659",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "reason": "Canvas BashService.listOutputs reads every result page sorted by timestamp, and no Canvas consumer polls BashOutput with order__gt. The reported duplicate order values affect reconstruction clients using that cursor; the current Canvas logs consumer does not use the affected order contract.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4658,+ "title": "[Bug]: Deferred initialization failure leaves a partially initialized runtime",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4658",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "feature_ids": [],+ "reason": "The report is transactionality/resource unwind in POST /api/init for a deferred warm-pool coordinator. Canvas neither calls that init endpoint nor owns its staged runtime publication, and no Canvas-observable failure is given in the report; this is a coordinator/Agent Server lifecycle contract before the UI attaches.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4629,+ "title": "[Bug]: Retired gemini-cli OAuth outranks a working API key, and the ACP Gemini pin is badly stale",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4629",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.acp-credentials"+ ],+ "basis": "live-reproduced",+ "reason": "An empty run-owned Gemini OAuth file produced the already-signed-in banner on two fresh ACP forms. Provider auth priority and version migration were not exercised.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "An empty run-owned Gemini OAuth file produced the already-signed-in banner on two fresh ACP forms. Provider auth priority and version migration were not exercised.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F13.acp-credentials/4629-empty-file-signed-in.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4613,+ "title": "[Bug]: OpenAI provider prefix stripped twice for namespaced model IDs in Responses API",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4613",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.custom-model"+ ],+ "basis": "known-issue",+ "reason": "A namespaced model such as openai/openai/example-model can have its inner provider prefix stripped again in Responses API and reach the provider as example-model. Reproduction needs an approved OpenAI-compatible endpoint serving the namespaced model and its exact stored API-mode settings; a loopback connection refusal cannot prove the outbound identifier.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4583,+ "title": "[Bug]: file editor insert concatenates onto last line when file has no trailing newline",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4583",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "A working model invoking file-editor insert at EOF on a file with no trailing newline.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4575,+ "title": "[Bug]: Agent terminal commands inherit server process priority",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4575",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "basis": "known-issue",+ "reason": "A model key and controlled bounded workload/resource envelope for a real Canvas terminal run are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4562,+ "title": "[Bug]: workspace files return 401 on private-LAN HTTP because the session cookie is rejected",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4562",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F08",+ "feature_ids": [+ "F08.files-load-error"+ ],+ "basis": "known-issue",+ "reason": "A plain-HTTP non-loopback same-origin browser deployment with real workspace-cookie observation.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4558,+ "title": "Encrypted LLM profiles are decrypted only on the conversation path — FallbackStrategy (and sub-agents) load them without the cipher",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4558",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "known-issue",+ "reason": "A working encrypted saved fallback profile and real primary-model failure that triggers fallback resolution.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4540,+ "title": "[Bug]: Tool calls return incorrect/weird syntax with Synthetic Provider",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4540",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "known-issue",+ "reason": "Synthetic provider credentials and a GLM/Kimi turn emitting the raw tool_call wrapper.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4537,+ "title": "[Bug]: TaskToolSet delegation holds the parent ConversationState lock for the whole sub-agent run, saturating the executor pool and freezing the conversation list on agent canvas UI",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4537",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "basis": "known-issue",+ "reason": "A working model with native TaskTool delegation holding the parent state lock during list reload.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4533,+ "title": "Conversation launch drops LLM api_key when the seeded `default` agent profile is active (litellm AuthenticationError)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4533",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "family": "F13",+ "feature_ids": [+ "F13.active-drives-conversation"+ ],+ "basis": "live-not-reproduced",+ "reason": "Two seeded-default UI launches preserved the dummy key in runtime state and reached loopback connection refusal. This verifies key propagation, not successful provider authentication.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @<owned-fixture>",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands api GET /api/agent-profiles/default --pick profile.llm_profile_ref",+ "control-openhands api GET /api/settings --pick llm_api_key_is_set",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation status fda5a4cd-9c39-4091-8b9a-5512f4c48642",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation status 950e5e83-f56d-4dbb-9aab-b2ccc78f1972",+ "python3 - <<'PY_KEY'\nimport json, os\nfrom pathlib import Path\nfor p in (Path(os.environ['OH_VERIFY_RUN']) / 'private').rglob('base_state.json'):\n d=json.loads(p.read_text())\n llm=d.get('agent',{}).get('llm',{})\n if llm.get('model')=='openai/gpt-4o': print({'conversation':p.parent.name,'model':llm.get('model'),'key_present':bool(llm.get('api_key'))})\nPY_KEY",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands doctor"+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F13.active-drives-conversation/default-key-diagnostic.json"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Two seeded-default UI launches preserved the dummy key in runtime state and reached loopback connection refusal. This verifies key propagation, not successful provider authentication.",+ "replay_setup": {+ "working_directory": "<checkout>",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "<owned-fixture>",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ }+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4533#issuecomment-6121790768"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4532,+ "title": "[Bug]: Reopening a confirmation-paused conversation drops the assistant tool_use message from the next request",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4532",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.confirmation-mode"+ ],+ "basis": "known-issue",+ "reason": "A real confirmation-paused action persisted across backend restart, then approval and a working provider call.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4511,+ "title": "[Bug]: `Message.to_chat_dict` emits `cache_control` when caching is disabled",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4511",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.schema-validation"+ ],+ "basis": "known-issue",+ "reason": "Disabling prompt caching can still emit cache_control on preexisting cache-marked tool messages. The report only supplies a Python serializer reproduction; a Canvas check needs the exact persisted/tool-message precondition and a provider request that exposes the cache markers, with approved model access.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4498,+ "title": "Incremental view path skips enforce_properties after condensation, leaving orphaned tool_results/tool_uses for the LLM",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4498",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "basis": "known-issue",+ "reason": "A real condensation splitting an action/result pair, followed by the next provider request.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4432,+ "title": "fix(mcp): pin fetch server runtime dependencies",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4432",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Fetch MCP runtime dependency pin request with no concrete Canvas reproduction or user-visible failure in the report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4382,+ "title": "[Bug]: ACP derived cost ignores cache/thought buckets, silently zeroes unknown models, and can inflate cost when UsageUpdate cost stalls",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4382",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "basis": "known-issue",+ "reason": "An authenticated ACP provider emitting cache/thought token usage or reset/stalled cumulative cost updates.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4363,+ "title": "InstallationManager.update() silently drops pinned refs (should reconcile, not force ref=None)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4363",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F19",+ "feature_ids": [+ "F19.refresh"+ ],+ "basis": "live-reproduced",+ "reason": "Completed Update changed persisted requested_ref from v1 to null. Visible version stayed 1.0.0 in this cache, so a default-HEAD jump was not observed.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "Completed Update changed persisted requested_ref from v1 to null. Visible version stayed 1.0.0 in this cache, so a default-HEAD jump was not observed.",+ "Exact CLI and arrange/restore steps are in the linked owning feature-map recipe."+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F19.refresh/4363-pinned-before.aria.txt",+ "backfill-18309-settings-worker/evidence/F19.refresh/pin-loss-diagnostic.json"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4093,+ "title": "ACP 0.11 drops Gemini model state from NewSessionResponse",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4093",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "basis": "known-issue",+ "reason": "With ACP Python 0.11, Gemini can lose the current/available model state from NewSessionResponse. Reproduction needs the affected ACP dependency and Gemini adapter plus an approved Gemini account; the profile's saved model field alone is not session model state.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4080,+ "title": "One unregistered event kind fails the entire conversation load (should degrade per-event)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4080",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "basis": "known-issue",+ "reason": "A copied persisted conversation with its original unregistered observation kind and missing tool-module registration.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4077,+ "title": "Streaming: correctness & resource-safety bugs in the token/delta pipeline",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4077",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F06",+ "feature_ids": [+ "F06.reload-mid-run"+ ],+ "basis": "known-issue",+ "reason": "A working stream interrupted/retried or delivered to a controlled stalled subscriber, with delta/state-order capture.",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4265,+ "title": "OpenHands Fork Audit - MEDIUM: Fork-delta maintenance risk — carried patches must survive upstream merges",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4265",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Zentra fork patch maintenance, outside this Canvas repository and supported runtime."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4264,+ "title": "OpenHands Fork Audit - HIGH: Stale duplicate build/lib/ directories under vendor/ could be shipped",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4264",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Zentra fork vendor build/lib packaging, outside this Canvas repository and runtime."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4263,+ "title": "OpenHands Fork Audit - HIGH: get_litellm_model_info makes unvalidated httpx.get call at LLM init",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4263",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Zentra fork air-gap/egress policy, outside this Canvas repository and supported runtime."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4260,+ "title": "[Bug]: Cloud Automation PR reviewer intermittently posts no review — SecretStr webhook serialization error (non-blocking) + 600s run ceiling killing stalled agent loop (blocker)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4260",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.run-linked-conversation",+ "F23.run-logs-modal"+ ],+ "basis": "known-issue",+ "reason": "Owned Cloud reviewer/webhook deployment, authorized disposable review target and model credentials are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4252,+ "title": "[Bug]: New added Global Skills dont get loaded into OpenHands (CLI install) and not available in WebUI",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4252",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "family": "F18",+ "feature_ids": [+ "F18.personal-skill-dirs"+ ],+ "basis": "live-not-reproduced",+ "reason": "A persistence-folder skill appeared in the UI and in available/activated skills in two actual initialized chats. This verifies discovery before the model reply, not historical Docker mounts.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @<owned-fixture>",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands fixture skill --name qa-legacy --trigger qa-legacy-ping",+ "mkdir -p \"$OH_VERIFY_RUN/private/skills\"",+ "mv \"$OH_VERIFY_RUN/private/home/.agents/skills/qa-legacy\" \"$OH_VERIFY_RUN/private/skills/qa-legacy\"",+ "control-openhands browser goto /skills",+ "control-openhands browser reload",+ "control-openhands browser count 'testid=skill-card-qa-legacy'",+ "control-openhands browser attr 'testid=skill-source-qa-legacy' title",+ "control-openhands conversation start --prompt \"qa-legacy-ping. Do not run tools; reply in one line.\" --timeout 60",+ "control-openhands conversation events 099f2448-1df7-4f7a-a702-7248dcf8a961 --grep qa-legacy --from-start --last 4",+ "control-openhands conversation start --prompt \"qa-legacy-ping. Do not run tools; reply in one line.\" --timeout 60",+ "control-openhands conversation events ca80b3a6-7882-4821-b591-a09c15b16210 --grep qa-legacy --from-start --last 4",+ "control-openhands browser screenshot --feature F18.personal-skill-dirs --name 4252-legacy-skill-loaded",+ "python3 - <<'PY_CLEAN'\nimport os,shutil\nfrom pathlib import Path\np=Path(os.environ['OH_VERIFY_RUN'])/'private/skills/qa-legacy'\nif p.exists():shutil.rmtree(p)\nPY_CLEAN",+ "control-openhands doctor"+ ],+ "evidence": [+ "backfill-18309-settings-worker/evidence/F18.personal-skill-dirs/4252-legacy-skill-loaded.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "A persistence-folder skill appeared in the UI and in available/activated skills in two actual initialized chats. This verifies discovery before the model reply, not historical Docker mounts.",+ "replay_setup": {+ "working_directory": "<checkout>",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "<owned-fixture>",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ }+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4252#issuecomment-6121791082"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4250,+ "title": "[Bug]: Workers AI broken",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4250",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.custom-model",+ "F10.default-used"+ ],+ "basis": "known-issue",+ "reason": "A Cloud Workers AI configuration can fail on context-window validation or be routed as OpenAI with an invalid Cloudflare key. Reproduction needs a disposable OpenHands Cloud account, its Workers AI endpoint/key and the exact saved model configuration; the report supplies no complete configuration steps.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4248,+ "title": "[Bug]: Missing required parameters for function 'execute_bash': {'security_risk'}",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4248",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "feature_ids": [],+ "reason": "This unspecified legacy uv CLI report names execute_bash/security_risk but gives no Canvas/Agent Server version, UI entry point or reproduction. The current Canvas launcher is not that SDK CLI; no affected supported Canvas runtime consumer is identified.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4731,+ "title": "Fix WebSocket client require('ws') breaking bundlers",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4731",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "TypeScript package bundler compatibility for require(ws); current Canvas browser event transport is separate and the frozen Canvas production build succeeds."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4247,+ "title": "[Bug]: LLM Provider NOT provided with LM Studio",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4247",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F10",+ "feature_ids": [+ "F10.custom-model"+ ],+ "basis": "known-issue",+ "reason": "An LM Studio configuration can fail with LLM Provider NOT provided although its models endpoint works. Reproduction needs the reachable LM Studio service and the exact model ID/provider-qualified config, which the report does not supply as text; the reported CLI/Docker SDK 1.16.1 installation is not this source stack.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4246,+ "title": "[Bug]:bserved Behavior: The agent remains idle. No visual feedback or error notification is shown in the interface.",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4246",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "family": "F17",+ "feature_ids": [+ "F17.agent-uses-server"+ ],+ "basis": "live-not-reproduced",+ "reason": "Two actual stdio timeouts left initialization after about 30 seconds, persisted the prompt/user message and reached loopback model error. This settles the current Local initialization loop only, not successful replies or historical Cloud/SSE behavior; sibling tools were lost.",+ "source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "live": {+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @<owned-fixture>",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands fixture mcp-server --name qa-settings-echo",+ "control-openhands api POST /api/settings/mcp/qa_settings_echo --write --data @<echo-config.json>",+ "python3 - <<'PY_SLOW'\nimport os\nfrom pathlib import Path\np=Path(os.environ['OH_VERIFY_RUN'])/'workspace/qa-settings-slow/server.mjs'\np.parent.mkdir(parents=True,exist_ok=True)\np.write_text('process.stdin.resume();setTimeout(()=>process.exit(0),90000);\\n')\nPY_SLOW",+ "control-openhands api POST /api/settings/mcp/qa_slow --write --data @<owned-fixture>",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation status c08b83fc-1753-4b70-8f6a-59396e9205ce",+ "control-openhands conversation events c08b83fc-1753-4b70-8f6a-59396e9205ce --kinds SystemPromptEvent,MessageEvent --from-start --last 3",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation status da523c3d-3dd1-4ce9-b592-90ba99477aae",+ "control-openhands conversation events da523c3d-3dd1-4ce9-b592-90ba99477aae --kinds SystemPromptEvent,MessageEvent --from-start --last 3",+ "control-openhands api DELETE /api/settings/mcp/qa_slow --write",+ "control-openhands api DELETE /api/settings/mcp/qa_settings_echo --write",+ "control-openhands doctor"+ ],+ "evidence": [],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Two actual stdio timeouts left initialization after about 30 seconds, persisted the prompt/user message and reached loopback model error. This settles the current Local initialization loop only, not successful replies or historical Cloud/SSE behavior; sibling tools were lost.",+ "replay_setup": {+ "working_directory": "<checkout>",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "<owned-fixture>",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ },+ "echo_config": {+ "file": "<echo-config.json>",+ "body": {+ "transport": "stdio",+ "command": "<Node executable returned by fixture mcp-server>",+ "args": [+ "<server.mjs returned by fixture mcp-server>"+ ],+ "enabled": true+ }+ },+ "slow_config": {+ "file": "<owned-fixture>",+ "body": {+ "transport": "stdio",+ "command": "<Node executable returned by fixture mcp-server>",+ "args": [+ "<OH_VERIFY_RUN>/workspace/qa-settings-slow/server.mjs"+ ],+ "enabled": true+ }+ }+ },+ "run_id": "backfill-18309-settings-worker"+ },+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4246#issuecomment-6121791434"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4245,+ "title": "[Bug]: Agent-Server Webhook Connection Failures Cause Container Crashes and Sandbox Connection Errors",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4245",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Legacy docker.openhands.dev/openhands/openhands v0 webhook deployment, absent from current Agent Canvas stack."+ },+ {+ "repository": "OpenHands/automation",+ "number": 565,+ "title": "[Bug]: 502 instead of 403 when the credential is not authorized for the X-Org-Id organization",+ "url": "https://github.com/OpenHands/automation/issues/565",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F25",+ "feature_ids": [+ "F25.cloud-org-rows",+ "F25.cloud-org-suspended"+ ],+ "basis": "known-issue",+ "reason": "Authenticated Cloud account/org-bound key and unauthorized-org fixture are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md"+ },+ {+ "repository": "OpenHands/automation",+ "number": 557,+ "title": "Prompt and plugin presets ignore the selected agent profile at run time",+ "url": "https://github.com/OpenHands/automation/issues/557",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.configuration",+ "F23.run-linked-conversation"+ ],+ "basis": "live-reproduced",+ "reason": "Saved bare qa-backfill-bare agent profile (tools=[], unique persona, closed-loopback qa-offline LLM), created prompt preset then PATCHed its selected agent_profile_id. Real UI Run now created a linked conversation whose launched_agent_profile is null, selected persona is absent and default tools are present; the selected profile was ignored before the deliberate closed-loopback LLM connection failure.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Selected profile on a prompt preset (`F23.configuration`, `F23.run-linked-conversation`).** Arrange a bare saved Agent Profile (`tools: []`, unique persona) referencing an offline LLM profile whose base URL is the closed loopback `http://127.0.0.1:9`; set zero retries and disable title generation. Create an owned active annual-cron prompt preset, PATCH its `agent_profile_id` to the bare profile, then click **Run now** in Canvas. Inspect the linked conversation's `launched_agent_profile`, tool names and persona without revealing secrets. Expected: the chosen profile governs that conversation. Known failure, reproduced 2026-10-12 against automation 1.19.3/Agent Server 1.54.0: provenance is null, the selected persona is absent, and default terminal/file-editor/task-tracker/finish tools replace the bare selection ([OpenHands/automation#557](https://github.com/OpenHands/automation/issues/557)). The deliberate loopback connection failure is incidental to this metadata check; no provider is reachable."+ ],+ "artifacts": [+ "evidence/F23.run-linked-conversation/profile-557.json",+ "evidence/F23.run-linked-conversation/profile-557.aria.txt"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/automation",+ "number": 514,+ "title": "fix(local): stop BashCommand when an automation run is cancelled",+ "url": "https://github.com/OpenHands/automation/issues/514",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.cancel-in-flight-run",+ "F23.activity-log",+ "F23.run-logs-modal"+ ],+ "basis": "live-reproduced",+ "reason": "Clicked Home owned worker Cancel run; toast Automation run cancelled and Activity Log CANCELLED. The python3 worker.py remains in this run launcher ancestry after cancellation, despite no provider or external worker calls.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Row menu, Run now and Cancel (`F03.home-row-menu`, `F03.cancel-in-flight-run`, `F03.home-automation-links`).** Run `control-openhands browser goto /` (the Manage link left the dashboard), then `control-openhands browser click '[data-testid^=\"running-automation-row-\"] >> has-text=QA_F03 Pong >> [data-testid^=\"running-automation-menu-\"]'` and `control-openhands browser snapshot '[data-testid^=\"running-automation-menu-panel-\"]'`: Run now, View, Edit, Turn off, separator, Pin to dashboard. `control-openhands browser click '[data-testid^=\"running-automation-run-\"]' --observe 'role=status'` observes `Automation dispatched`. Reopen the menu: `Run now [disabled]` and **Cancel run**. `control-openhands browser click '[data-testid^=\"running-automation-cancel-\"]'` and `control-openhands browser toasts` reads `Automation run cancelled`; `control-openhands api GET /api/automation/v1/<automation-id>/runs` (id from `browser testids --filter running-automation-row`) shows `CANCELLED`, and after `browser reload` the row reads `Just now · Cancelled`. Pong now has a run: `control-openhands browser click '[data-testid^=\"running-automation-activity-\"] >> role=link' --expect-url '/automations/[^/?]+\\?run='` opens it; return with `control-openhands browser goto /`. Known failure (reproduced 2026-10-12, [OpenHands/automation#514](https://github.com/OpenHands/automation/issues/514)): a bounded script run cancelled from this Home menu shows `Automation run cancelled` and `Cancelled` in the row and Activity Log, but its launcher-owned `python3 worker.py` process remains running after cancellation. This was reproduced in `backfill-18309-integrations-worker` on `fa1f724ab5737a17453ad6081129eb91c5a423c6`, SDK 1.54.0 and automation 1.19.3: package a run-owned `worker.py` that prints one tick and sleeps 2 seconds for 90 iterations; upload its tarball with `control-openhands api POST '/api/automation/v1/uploads?name=qa-backfill-sleep' --write --data-file <bounded-sleep-tarball> --content-type application/gzip`, then create an active scripted automation through `api POST /api/automation/v1 --write --data '@<fixture-json>'` with `entrypoint: python3 worker.py`, the returned `tarball_path`, a UTC `0 0 1 1 *` cron trigger and timeout 240. Open `/automations/<sleep-id>`, click `role=button[name=\"Run now\"]`, wait for the actual worker to start, return Home and click `testid=running-automation-menu-<sleep-id>` then `testid=running-automation-cancel-<sleep-id>`. Read-only process checks restricted to this launcher's descendants still found the worker after the cancellation UI settled. Evidence: `F03.cancel-in-flight-run/local-worker-cancel-514.png`, `F03.cancel-in-flight-run/owned-worker-post-cancel.json` and `F23.activity-log/cancelled-514.aria.txt`; the worker is bounded to exit without external calls. Let it exit, then delete the temporary automation with `control-openhands api DELETE /api/automation/v1/<sleep-id> --write`."+ ],+ "artifacts": [+ "evidence/F03.cancel-in-flight-run/owned-worker-post-cancel.json",+ "evidence/F23.activity-log/cancelled-514.aria.txt"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md"+ },+ {+ "repository": "OpenHands/automation",+ "number": 472,+ "title": "Release conversation subjects when runs are cancelled",+ "url": "https://github.com/OpenHands/automation/issues/472",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Automation subject ownership and cancellation retry contract; Canvas does not submit subject-turn idempotency keys."+ },+ {+ "repository": "OpenHands/automation",+ "number": 470,+ "title": "fix: retry unsuccessful idempotent subject turns",+ "url": "https://github.com/OpenHands/automation/issues/470",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "reason": "Automation subject-turn retry/idempotency API; Canvas does not submit subject turns or idempotency keys in its UI."+ },+ {+ "repository": "OpenHands/automation",+ "number": 418,+ "title": "[pr-review-bot] Improvements and Bug Fixes for the pr-review-bot",+ "url": "https://github.com/OpenHands/automation/issues/418",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "PR-review-bot merge gating, review attribution and operational audit; no concrete mapped Canvas UI behavior."+ },+ {+ "repository": "OpenHands/automation",+ "number": 347,+ "title": "Run summaries are only readable one automation at a time, and the SQLite engine's undeclared pool turns the resulting fan-out into 30s stalls + HTTP 500",+ "url": "https://github.com/OpenHands/automation/issues/347",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F21",+ "feature_ids": [+ "F21.card-grid",+ "F21.overview-tiles"+ ],+ "basis": "known-issue",+ "reason": "The owned 25-automation local dashboard completed run summaries with HTTP 200 and no 30 s stall on automation 1.19.3. This narrow live probe does not account for the issue’s missing batch-summary API/pool configuration contracts or withheld original lock workload. A failing dashboard recipe needs the original DB load/lock setup and redacted service-log/HAR timing evidence, linked to Canvas #16559.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "The owned 25-automation local dashboard completed run summaries with HTTP 200 and no 30 s stall on automation 1.19.3. This narrow live probe does not account for the issue’s missing batch-summary API/pool configuration contracts or withheld original lock workload. A failing dashboard recipe needs the original DB load/lock setup and redacted service-log/HAR timing evidence, linked to Canvas #16559."+ ],+ "artifacts": [+ "evidence/F21.card-grid/scale-347.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md"+ },+ {+ "repository": "OpenHands/automation",+ "number": 343,+ "title": "setup_script_path is validated, stored and tested — but never read; only a root setup.sh ever runs",+ "url": "https://github.com/OpenHands/automation/issues/343",+ "ready_for_dev_at_snapshot": false,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F22.setup-tarball-create",+ "F23.script-section",+ "F23.run-logs-modal"+ ],+ "basis": "live-reproduced",+ "reason": "Uploaded owned tarball scripts/setup.sh writes .qa-setup-marker, main.py requires marker; persisted setup_script_path scripts/setup.sh. UI Run now genuine BashOutput exits1 with QA_NESTED_SETUP_WAS_SKIPPED, proving nested setup ignored.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Custom nested setup (`F23.script-section`, `F23.run-logs-modal`; creation in F22).** Arrange a reviewed tarball containing `scripts/setup.sh` that writes `.qa-setup-marker` and `main.py` that asserts that marker exists. Upload with `api POST /api/automation/v1/uploads?name=qa-nested-setup --write --data-file <tarball> --content-type application/gzip`; create with `entrypoint: python3 main.py`, `setup_script_path: scripts/setup.sh`, active state and yearly cron. Open the detail page and click **Run now**; correlate its `bash_command_id` with `api GET '/api/bash/bash_events/search?command_id__eq=<id>&limit=100'`. Expected: the configured nested setup runs before the entrypoint. Known failure, reproduced 2026-10-12 against automation 1.19.3: the real BashOutput exits 1 with the marker assertion, proving the setup step was skipped ([OpenHands/automation#343](https://github.com/OpenHands/automation/issues/343)). Only an owned marker is written; no external service or model is used."+ ],+ "artifacts": [+ "evidence/F23.activity-log/nested-status-proof.json"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/automation",+ "number": 286,+ "title": "Background scheduler, dispatcher and watchdog start once per uvicorn worker instead of once per deployment",+ "url": "https://github.com/OpenHands/automation/issues/286",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling"+ ],+ "basis": "known-issue",+ "reason": "Owned multi-worker/replica automation deployment and authenticated Cloud sandbox verification context are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/automation",+ "number": 246,+ "title": "automation",+ "url": "https://github.com/OpenHands/automation/issues/246",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F21",+ "feature_ids": [+ "F21.backend-unavailable",+ "F21.list-error"+ ],+ "basis": "known-issue",+ "reason": "The exact 404 URL, user action, version and local launcher/service configuration are missing from the report.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md"+ },+ {+ "repository": "OpenHands/automation",+ "number": 219,+ "title": "Automation sandbox creation can evict a user's own live session; the graceful-skip path is dead",+ "url": "https://github.com/OpenHands/automation/issues/219",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling"+ ],+ "basis": "known-issue",+ "reason": "Owned disposable Cloud account at its sandbox cap with an interactive fixture is unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ },+ {+ "repository": "OpenHands/automation",+ "number": 200,+ "title": "[post-release] Harden KV request models (extra=forbid) + fix missing-key inconsistency + version in write responses",+ "url": "https://github.com/OpenHands/automation/issues/200",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "reason": "Automation KV REST request strictness and response contracts; no Canvas KV editor or consumer of these writes."+ },+ {+ "repository": "OpenHands/automation",+ "number": 113,+ "title": "Bug: UI says conversations not started, when the conversation did start",+ "url": "https://github.com/OpenHands/automation/issues/113",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "family": "F23",+ "feature_ids": [+ "F23.run-linked-conversation"+ ],+ "basis": "known-issue",+ "reason": "Affected automation/run/conversation ids, deployment version and genuine plugin-run credentials are unavailable.",+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md"+ }+ ]+}diff --git a/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.md b/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.mdnew file mode 100644index 0000000000..1257438ca1--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.md@@ -0,0 +1,440 @@+# Existing-bug backfill — 2026-10-12++I am an AI Agent (Codex) helping Engel Nyst (@enyst) with OpenHands project work.++**Verdict: complete accounting of the frozen inventory.**+Mapped means a tracking record. It does not claim every mapped issue reproduced on this target. Expected results stay intact.++Target: `main@fa1f724ab5737a17453ad6081129eb91c5a423c6`. The maintenance baseline remains `main@553d4519113192a80fb18376fdb6c1e39fa06387`; no baseline change is proposed.++[Full per-issue ledger](2026-10-12-open-bugs.json). [Live check ledger](2026-10-12-open-bugs-checks.json). [Outcome and label workflow](../bug-backfill.md).++## Inventory and outcomes++| Repository | Open bug inventory | Mapped | Not reproduced | Out of scope | Pending |+| --- | ---: | ---: | ---: | ---: | ---: |+| OpenHands/OpenHands | 208 | 172 | 10 | 26 | 0 |+| OpenHands/software-agent-sdk | 141 | 93 | 7 | 41 | 0 |+| OpenHands/automation | 13 | 9 | 0 | 4 | 0 |++Inventories exclude pull requests. Every SDK and automation bug was inspected for a concrete current Canvas consumer; unused client paths and backend-only contracts have scope reasons. Inventory timestamps are separate from the Git target:++- OpenHands/OpenHands: `2026-10-12T04:49:30.813663+00:00` UTC.+- OpenHands/software-agent-sdk: `2026-10-12T04:50:37.186198+00:00` UTC.+- OpenHands/automation: `2026-10-12T04:50:33.550521+00:00` UTC.++`ready-for-dev` prioritized live work. All 208 main-repository bugs are accounted for, including 132 unmarked reports (76 marked). Readiness, priority and existing `reproduced` labels were preserved. The three colored outcome labels are applied only in OpenHands/OpenHands; cross-repository accounting stays in this report and qualified map links.++## Mapped evidence++| Basis | Count |+| --- | ---: |+| existing-note | 33 |+| known-issue | 174 |+| live-reproduced | 67 |++Existing notes keep their historical dates. New known failures have current live evidence. Other known issues name the missing account, platform, model success path, original fixture or reproduction detail. A layout-only observation is marked as such. This issue pass does not claim a full live audit of all 747 feature-map IDs.++## Runtime and checks++The product was built with `npm ci --ignore-scripts --no-audit --no-fund` and `npm run build:app`. Native macOS arm64 (Darwin 27.0.0), Node 25.9.0, Playwright 1.63.0 with actual HeadlessChrome 153.0.8010.12, app default theme/fonts; Agent Server/SDK 1.54.0 and automation 1.19.3. Build identity: `d465c70f971b4d8e`. The worktree began at the target revision; only skill files changed. No BASE runtime was run, so new reproductions establish presence on TARGET, without a claim about when they began.++Each worker used its own run, state, ports and evidence ledger. Runs overlapped only in isolated stacks after checking available/reclaimable memory; idle coordinator runs were stopped to release resources. Codex (GPT-6) performed the pass. No real model credentials or paid inference were used. Disposable initialization/error fixtures used refused or bounded loopback endpoints. One Basic edit cleared its saved base URL before validation and sent a dummy credential to OpenAI, which rejected it before inference. That attempt is a setup mistake, not product proof; the owned server fallback URLs were protected before replay. Missing Cloud, Windows, Docker, Electron and successful model prerequisites remain explicit in notes.++| Run | UTC start → stop | Ports: ingress / agent / automation / frontend / editor | Checks |+| --- | --- | --- | --- |+| `backfill-18309-coordinator` | 2026-10-12T05:31:38.273Z → 2026-10-12T05:34:29.215Z | 18820 / 18821 / 18822 / 18823 / 19821 | {} |+| `backfill-18309-coordinator-checks` | 2026-10-12T06:05:36.380Z → 2026-10-12T09:31:53.387Z | 18840 / 18841 / 18842 / 18843 / 19841 | {'pass': 1, 'fail': 3} |+| `backfill-18309-integrations-worker` | 2026-10-12T05:33:54.431Z → 2026-10-12T06:23:39.105Z | 18810 / 18811 / 18812 / 18813 / 19811 | {'fail': 20, 'pass': 7, 'blocked': 5} |+| `backfill-18309-local-secondary` | 2026-10-12T09:12:20.264Z → 2026-10-12T09:31:53.555Z | 18860 / 18861 / 18862 / 18863 / 19861 | {} |+| `backfill-18309-settings-rechecks` | 2026-10-12T09:31:36.011Z → 2026-10-12T10:33:01.788Z | 18820 / 18821 / 18822 / 18823 / 19821 | {'pass': 3, 'fail': 7} |+| `backfill-18309-settings-worker` | 2026-10-12T05:28:25.778Z → 2026-10-12T08:05:56.859Z | 18800 / 18801 / 18802 / 18803 / 19801 | {'fail': 30, 'pass': 9, 'blocked': 1} |+| `backfill-18309-surface` | 2026-10-12T04:54:13.393Z → 2026-10-12T04:57:49.620Z | 18800 / 18801 / 18802 / 18803 / 19801 | {'pass': 1} |+| `backfill-18309-surface-rechecks` | 2026-10-12T09:12:56.383Z → 2026-10-12T10:10:24.208Z | 18810 / 18811 / 18812 / 18813 / 19811 | {'pass': 2, 'fail': 4, 'blocked': 1} |+| `backfill-18309-surface-worker` | 2026-10-12T04:58:12.826Z → 2026-10-12T05:21:42.086Z | 18800 / 18801 / 18802 / 18803 / 19801 | {'fail': 11} |++Harness extension: `browser network-state offline|online|status` changes real browser-context connectivity, without response interception. The focused TDD test failed before implementation; the final 57 harness tests passed; eight Python fixture tests also passed using the installed SDK environment (which supplies httpx). Live readback showed `navigator.onLine` true → false → true and an uncached request 200 → TypeError → 200. A real reconnect also reproduced unsaved settings loss; the original expected behavior remains unchanged.++[Reviewed recording](evidence/network-state-and-refetch.mp4) and [GIF](evidence/network-state-and-refetch.gif); [before refetch](evidence/16948-before-refetch.png) and [after refetch](evidence/16948-after-refetch.png). These show a harness capability and an existing product failure, not a product fix.++The new WebSocket observer captured a real same-page backend restart: with three stored events, the reconnect cursor advanced from `-1` to `2`, and the recovery marker appeared once. [Recovered Canvas](evidence/17619-reconnected.png) and [URL-free lifecycle summary](evidence/17619-socket-cursor.json) demonstrate the observer without publishing frame bodies or session URLs. This scoped result does not establish zero duplicate protocol frames in every timing window. Bounded tracked fixtures arrange a non-responding model endpoint, ZIP download overlap and a secondary Agent Server sharing disposable conversation storage; they do not replace model output.++- `control-openhands map check`: ok.+- `control-openhands map coverage`: ok: 31 routes and 29 feature directories accounted for.+- `control-openhands map testids`: ok: 1037 cited, 0 unresolved.+- `focused harness tests`: 57/57 passed (focused Vitest suite).+- `bounded fixture tests`: 8/8 passed (installed SDK Python with httpx; bounded ZIP/transport and process-group cleanup).+- `git diff --check`: clean.++## Per-issue accounting++| Issue | Outcome | Map/check | Basis or scope reason |+| --- | --- | --- | --- |+| [OpenHands/OpenHands#18303](https://github.com/OpenHands/OpenHands/issues/18303) | mapped | [F10.subscription](../feature-map/F10-llm-profiles.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#18302](https://github.com/OpenHands/OpenHands/issues/18302) | mapped | [F01.onboarding-repeat-endpoint](../feature-map/F01-first-run-and-sign-in.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#18301](https://github.com/OpenHands/OpenHands/issues/18301) | mapped | [F26.session-key-rotated / F26.session-key-persist](../feature-map/F26-runtime-variants.md) | Wrote and read an owned Automation KV marker (HTTP 201/200), then control-openhands restart --rotate-key with no AUTOMATION_KV_SECRET override. Fresh-auth KV GET now returns HTTP 500 instead of the prior marker. |+| [OpenHands/OpenHands#18299](https://github.com/OpenHands/OpenHands/issues/18299) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Stopped only the owned automation service, backed up its DB, reset alembic_version from 032 to 028 while consecutive_auth_failures exists, restarted the real launcher. Migration duplicate-column signature and automation unavailability occur while launcher stays alive. Private original DB restored afterward. |+| [OpenHands/OpenHands#18287](https://github.com/OpenHands/OpenHands/issues/18287) | mapped | [F18.card / F18.facets](../feature-map/F18-skills.md) | After filtering a personal/bundled name collision, the summary said one result while two then three stale skill-creator cards remained. |+| [OpenHands/OpenHands#18234](https://github.com/OpenHands/OpenHands/issues/18234) | mapped | [F03.home-run-outcome-health](../feature-map/F03-home.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#18233](https://github.com/OpenHands/OpenHands/issues/18233) | mapped | [F25.selector-dropdown](../feature-map/F25-backends-and-cloud.md) | Historical reproduced note dated 2026-10-09 already describes closed-menu Escape clearing the active backend display; source Dropdown and backend selector still own the path. |+| [OpenHands/OpenHands#18210](https://github.com/OpenHands/OpenHands/issues/18210) | mapped | [F21.manifest-404](../feature-map/F21-automations-dashboard.md) | Existing missing-manifest recipe and note describe the Home link/toasts; preserved source revision and prior evidence date. |+| [OpenHands/OpenHands#18161](https://github.com/OpenHands/OpenHands/issues/18161) | mapped | [F17.credential-probe](../feature-map/F17-mcp-servers.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#18160](https://github.com/OpenHands/OpenHands/issues/18160) | mapped | [F26.frontend-only-returning](../feature-map/F26-runtime-variants.md) | Existing reproduced 2026-10-08 recovery-gate hydrator/toast note preserved. |+| [OpenHands/OpenHands#18155](https://github.com/OpenHands/OpenHands/issues/18155) | mapped | [F04.show-archived-unarchive](../feature-map/F04-conversation-list.md) | A Cloud sandbox or local runtime that becomes MISSING and non-resumable; ordinary local pause remains resumable. |+| [OpenHands/OpenHands#18136](https://github.com/OpenHands/OpenHands/issues/18136) | mapped | [F07.cloud-only](../feature-map/F07-conversation-page.md) | An authenticated Cloud sandbox/repository provider and a second fresh browser to compare server repository metadata. |+| [OpenHands/OpenHands#18117](https://github.com/OpenHands/OpenHands/issues/18117) | mapped | [F06.thinking](../feature-map/F06-agent-activity.md) | A live model that emits a leading inline think block on a tool-call turn, plus its finalized ActionEvent. |+| [OpenHands/OpenHands#18108](https://github.com/OpenHands/OpenHands/issues/18108) | mapped | [F08.files-tree](../feature-map/F08-workspace-files-and-changes.md) | An authenticated Enterprise Cloud sandbox, no repository selected, with files outside its default working directory. |+| [OpenHands/OpenHands#18063](https://github.com/OpenHands/OpenHands/issues/18063) | mapped | [F05.overflow-menu](../feature-map/F05-composer.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#18062](https://github.com/OpenHands/OpenHands/issues/18062) | mapped | [F21.run-now-error / F23.run-now-error](../feature-map/F23-automation-detail.md) | Existing dashboard and detail stale-automation notes and recipes preserved; no fresh reproduction claimed. |+| [OpenHands/OpenHands#18029](https://github.com/OpenHands/OpenHands/issues/18029) | mapped | [F25.locked-cloud](../feature-map/F25-backends-and-cloud.md) | Affected locked-to-cloud deployment, authenticated Cloud session and stale-document/alternate-origin evidence are unavailable. |+| [OpenHands/OpenHands#17959](https://github.com/OpenHands/OpenHands/issues/17959) | mapped | [F22.templates-search / F22.template-card](../feature-map/F22-automation-creation.md) | Both no-result template search and optional-integration pill are already recorded; preserved existing notes. |+| [OpenHands/OpenHands#17955](https://github.com/OpenHands/OpenHands/issues/17955) | mapped | [F10.create-prefill](../feature-map/F10-llm-profiles.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#17953](https://github.com/OpenHands/OpenHands/issues/17953) | mapped | [F25.edit-escape / F25.shared-conversation-view / F25.cloud-device-flow](../feature-map/F25-backends-and-cloud.md) | All three reported failures already have existing map notes and drive recipes. |+| [OpenHands/OpenHands#17949](https://github.com/OpenHands/OpenHands/issues/17949) | mapped | [F26.host-bind-lan-optin / F26.cli-port-in-use / F26.backend-only](../feature-map/F26-runtime-variants.md) | All three launcher defects already mapped; retained reproduced 2026-10-08 flag note. |+| [OpenHands/OpenHands#17948](https://github.com/OpenHands/OpenHands/issues/17948) | mapped | [F23.edit-time-cleared / F23.run-no-conversation / F23.not-found / F23.tarball](../feature-map/F23-automation-detail.md) | All four automation-detail defects already recorded in Gotchas and corresponding recipes. |+| [OpenHands/OpenHands#17946](https://github.com/OpenHands/OpenHands/issues/17946) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Existing npm tmux-isolation known issue preserved; verification launcher intentionally supplies isolation and is not a product-fix claim. |+| [OpenHands/OpenHands#17941](https://github.com/OpenHands/OpenHands/issues/17941) | mapped | [F18.toggle / F18.toggle-error](../feature-map/F18-skills.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#17932](https://github.com/OpenHands/OpenHands/issues/17932) | mapped | [F15.save-errors](../feature-map/F15-agent-behavior-settings.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#17927](https://github.com/OpenHands/OpenHands/issues/17927) | mapped | [F16.save-states](../feature-map/F16-application-settings.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#17922](https://github.com/OpenHands/OpenHands/issues/17922) | mapped | [F17.install-unsupported](../feature-map/F17-mcp-servers.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#17920](https://github.com/OpenHands/OpenHands/issues/17920) | mapped | [F06.confirmation-mode](../feature-map/F06-agent-activity.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#17919](https://github.com/OpenHands/OpenHands/issues/17919) | mapped | [F04.delete-all](../feature-map/F04-conversation-list.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#17900](https://github.com/OpenHands/OpenHands/issues/17900) | mapped | [F16.keyboard](../feature-map/F16-application-settings.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#17899](https://github.com/OpenHands/OpenHands/issues/17899) | mapped | [F16.git-identity](../feature-map/F16-application-settings.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#17898](https://github.com/OpenHands/OpenHands/issues/17898) | not-reproduced | [F16.analytics](https://github.com/OpenHands/OpenHands/issues/17898#issuecomment-6121789760) | The hard-DNT build sent zero PostHog requests before and after opt-in/Save/reload, and created no PostHog storage key. Consent was restored off. |+| [OpenHands/OpenHands#17872](https://github.com/OpenHands/OpenHands/issues/17872) | out-of-scope | — | Mock-LLM E2E CI failures and stale test selectors; test harness coverage rather than a proved Canvas user feature failure. |+| [OpenHands/OpenHands#17871](https://github.com/OpenHands/OpenHands/issues/17871) | out-of-scope | — | Live ACP E2E harness compile-time constant setup; test harness startup rather than the production Canvas path. |+| [OpenHands/OpenHands#17806](https://github.com/OpenHands/OpenHands/issues/17806) | not-reproduced | [F10.edit](https://github.com/OpenHands/OpenHands/issues/17806#issuecomment-6122466133) | Two Advanced and one protected Basic real model-only saves preserved the untouched hidden dummy key (api_key_set=true); subsequent actual conversation switches materialized the changed model with a populated runtime key. Basic intentionally removed base_url, so the owned server default OPENAI_BASE_URL/OPENAI_API_BASE were set to refused loopback before its successful Save. Scope is current Local same-provider key preservation/propagation; provider changes, Cloud and successful authentication remain untested. |+| [OpenHands/OpenHands#17803](https://github.com/OpenHands/OpenHands/issues/17803) | mapped | [F11.edit / F11.agent-uses-connection](../feature-map/F11-provider-connections.md) | Two real connection Base URL edits persisted /v2 then /v3, but two fresh Canvas-created conversations still materialized the activated /v1 URL. The stored connection, settings snapshot and run-owned runtime state were read separately. This proves stale URL resolution before model success; all URLs were refused loopback. |+| [OpenHands/OpenHands#17801](https://github.com/OpenHands/OpenHands/issues/17801) | mapped | [F10.create-prefill / F10.edit](../feature-map/F10-llm-profiles.md) | Add and Edit selected All twice for a profile with a custom base URL and temperature 0.4, rather than the requested Basic starting tab. |+| [OpenHands/OpenHands#17788](https://github.com/OpenHands/OpenHands/issues/17788) | mapped | [F05.attach-files](../feature-map/F05-composer.md) | A Cloud sandbox whose cached runtime URL/key are blank while the fresh runtime lookup is already ready. |+| [OpenHands/OpenHands#17785](https://github.com/OpenHands/OpenHands/issues/17785) | mapped | [F10.create-validation](../feature-map/F10-llm-profiles.md) | Saving DeepSeek-v4-flash on Windows Desktop can fail with missing cache_creation_tokens instead of saving. The reported path needs a Windows Desktop 1.24.0 environment with its SDK dependency resolution and an approved DeepSeek key; this macOS source run cannot establish that result. |+| [OpenHands/OpenHands#17783](https://github.com/OpenHands/OpenHands/issues/17783) | mapped | [F20.git-tree-url / F20.install-error](../feature-map/F20-canvas-apps.md) | Executed both malformed ref and path tree URLs in the live install modal; each raised URI malformed, no install mutation or toast, modal remains. Reviewed local demo control installs and enables. |+| [OpenHands/OpenHands#17774](https://github.com/OpenHands/OpenHands/issues/17774) | mapped | [F27.planner-empty / F27.planner-plan](../feature-map/F27-workspace-tools.md) | Native Windows local Agent Server and drive-letter workspace fixtures are unavailable. |+| [OpenHands/OpenHands#17773](https://github.com/OpenHands/OpenHands/issues/17773) | mapped | [F08.files-tree](../feature-map/F08-workspace-files-and-changes.md) | A Windows local Agent Server using cmd.exe; this run is macOS. |+| [OpenHands/OpenHands#17763](https://github.com/OpenHands/OpenHands/issues/17763) | mapped | [F26.docker-image](../feature-map/F26-runtime-variants.md) | An owned Docker container/daemon is unavailable. |+| [OpenHands/OpenHands#17710](https://github.com/OpenHands/OpenHands/issues/17710) | mapped | [F14.delete / F14.edit](../feature-map/F14-secrets.md) | On the Cloud backend, deleting a secret reportedly adds a suffixed duplicate and leaves the original. Reproduction needs a disposable writable OpenHands Cloud account at the affected service; Local secret CRUD exercises a different transport. |+| [OpenHands/OpenHands#17706](https://github.com/OpenHands/OpenHands/issues/17706) | mapped | [F10.default-used](../feature-map/F10-llm-profiles.md) | A freshly resolved npm-launcher environment can reject openhands/claude-* conversations because LiteLLM sends prompt_cache_key. Reproduction needs an approved OpenHands proxy key with an Anthropic model and the affected fresh transitive dependency resolution; a keyless initialization cannot prove provider acceptance. |+| [OpenHands/OpenHands#17704](https://github.com/OpenHands/OpenHands/issues/17704) | mapped | [F23.script-section](../feature-map/F23-automation-detail.md) | Owned UTF-8 USTAR bundle café.py/main.py, inactive automation python3 café.py: actual Script header renders café.py and no Entrypoint badge; ASCII control untouched. |+| [OpenHands/OpenHands#17657](https://github.com/OpenHands/OpenHands/issues/17657) | out-of-scope | — | Release-ready workflow permission/startup failure; repository CI tooling. |+| [OpenHands/OpenHands#17623](https://github.com/OpenHands/OpenHands/issues/17623) | mapped | [F26.helm-chart](../feature-map/F26-runtime-variants.md) | Helm and a disposable Kubernetes cluster are unavailable. |+| [OpenHands/OpenHands#17622](https://github.com/OpenHands/OpenHands/issues/17622) | out-of-scope | — | Node requirement documentation drift; documentation only. |+| [OpenHands/OpenHands#17621](https://github.com/OpenHands/OpenHands/issues/17621) | mapped | [F05.context-window-meter](../feature-map/F05-composer.md) | A real summarizing model and conversation whose condensation takes longer than 60 seconds. |+| [OpenHands/OpenHands#17620](https://github.com/OpenHands/OpenHands/issues/17620) | out-of-scope | — | Fresh-clone typecheck generation order; contributor build tooling. |+| [OpenHands/OpenHands#17619](https://github.com/OpenHands/OpenHands/issues/17619) | not-reproduced | [F06.reconnect](https://github.com/OpenHands/OpenHands/issues/17619#issuecomment-6122465227) | The initial session socket used after_seq=-1. Three genuine run:false user messages yielded events/count=7; the same open page reconnected after an owned backend restart with after_seq=6. A recovery message appeared once. A recorded repeat advanced -1 to2 with count3. No frame bodies were captured, so this does not quantify server replayed payloads or cost. |+| [OpenHands/OpenHands#17577](https://github.com/OpenHands/OpenHands/issues/17577) | mapped | [F10.model-picker](../feature-map/F10-llm-profiles.md) | Basic OpenRouter discarded the off-catalog model ID on blur in two trials; the model field became empty. |+| [OpenHands/OpenHands#17567](https://github.com/OpenHands/OpenHands/issues/17567) | mapped | [F08.diff-view-modes](../feature-map/F08-workspace-files-and-changes.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#17566](https://github.com/OpenHands/OpenHands/issues/17566) | mapped | [F27.terminal-history-reload](../feature-map/F27-workspace-tools.md) | Existing terminal history failure reproduced 2026-10-06 is preserved; no available model credential for a new terminal command. |+| [OpenHands/OpenHands#17564](https://github.com/OpenHands/OpenHands/issues/17564) | not-reproduced | [F21.phone](https://github.com/OpenHands/OpenHands/issues/17564#issuecomment-6121791800) | With 25 inert automation fixtures, the dashboard heading measures 347 px at 390 px and 277 px at 320 px, at 28 px height. Actions stack below it and no horizontal page overflow appears. This exact current dashboard layout does not reproduce the reported 52 px squeeze. |+| [OpenHands/OpenHands#17563](https://github.com/OpenHands/OpenHands/issues/17563) | mapped | [F23.phone / F23.edit-close](../feature-map/F23-automation-detail.md) | Existing mobile clipping and Escape notes preserved. |+| [OpenHands/OpenHands#17559](https://github.com/OpenHands/OpenHands/issues/17559) | mapped | [F06.load-older-history](../feature-map/F06-agent-activity.md) | A real persisted history page boundary splitting an action/observation or ACP started/completed pair; dummy-key errors do not produce these tool pairs. |+| [OpenHands/OpenHands#17546](https://github.com/OpenHands/OpenHands/issues/17546) | mapped | [F26.cli-port-in-use](../feature-map/F26-runtime-variants.md) | Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. It provides no automatic frontend fallback, and the ingress override does not change that internal port. |+| [OpenHands/OpenHands#17520](https://github.com/OpenHands/OpenHands/issues/17520) | mapped | [F10.delete](../feature-map/F10-llm-profiles.md) | Two deletes returned 200 while title-cleanup PATCH returned 500; the UI kept the deleted row although its API returned 404. Temporary own-file immutable flags were restored immediately and doctor passed. |+| [OpenHands/OpenHands#17508](https://github.com/OpenHands/OpenHands/issues/17508) | mapped | [F09.update-available](../feature-map/F09-settings-shell.md) | Isolated old/new package installs with the old process still serving, or the original stale bundle/cache capture after update. |+| [OpenHands/OpenHands#17497](https://github.com/OpenHands/OpenHands/issues/17497) | mapped | [F04.card](../feature-map/F04-conversation-list.md) | The original corrupted persisted conversation or a deterministic corruption fixture after model-switch failure and fork. |+| [OpenHands/OpenHands#17494](https://github.com/OpenHands/OpenHands/issues/17494) | mapped | [F07.export-transcript](../feature-map/F07-conversation-page.md) | An authenticated ACP session that emits a titleless tool call, then Markdown and HTML export. |+| [OpenHands/OpenHands#17492](https://github.com/OpenHands/OpenHands/issues/17492) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model emitting failed browser, glob, grep, planning-editor and file-editor observations with is_error=true. |+| [OpenHands/OpenHands#17490](https://github.com/OpenHands/OpenHands/issues/17490) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model producing a completed terminal command with nonzero exit_code and is_error=false. |+| [OpenHands/OpenHands#17485](https://github.com/OpenHands/OpenHands/issues/17485) | out-of-scope | — | The reported persistent shared bash-events correlation consumer was removed. Current useBashCommandRunner uses conversation-scoped BashClient.executeCommand. |+| [OpenHands/OpenHands#17476](https://github.com/OpenHands/OpenHands/issues/17476) | mapped | [F05.context-window-meter](../feature-map/F05-composer.md) | A real planning sub-conversation and main conversation with distinct nonzero usage stats. |+| [OpenHands/OpenHands#17471](https://github.com/OpenHands/OpenHands/issues/17471) | mapped | [F05.attach-files](../feature-map/F05-composer.md) | Same-file re-pick after removing its chip leaves attachment count zero. |+| [OpenHands/OpenHands#17470](https://github.com/OpenHands/OpenHands/issues/17470) | mapped | [F26.docker-image](../feature-map/F26-runtime-variants.md) | Affected Apple Silicon Colima ARM container environment and Docker daemon are unavailable. |+| [OpenHands/OpenHands#17436](https://github.com/OpenHands/OpenHands/issues/17436) | mapped | [F22.setup-review-confirm](../feature-map/F22-automation-creation.md) | A real LLM key and the reported legacy release/template pair are unavailable. |+| [OpenHands/OpenHands#17429](https://github.com/OpenHands/OpenHands/issues/17429) | mapped | [F05.slash-btw](../feature-map/F05-composer.md) | The exact pending tool/streaming state that triggers the generator-pickle error, plus a working model for the side question. |+| [OpenHands/OpenHands#17426](https://github.com/OpenHands/OpenHands/issues/17426) | mapped | [F13.set-active](../feature-map/F13-agent-profiles.md) | The active agent profile badge says Default while its menu says Set as active, on two openings. |+| [OpenHands/OpenHands#17391](https://github.com/OpenHands/OpenHands/issues/17391) | mapped | [F10.subscription](../feature-map/F10-llm-profiles.md) | ChatGPT Business subscription login can be refused or use a fallback model with a 404. Completing the entitlement and selected-model checks needs an approved ChatGPT Business workspace login and the reported Docker subscription runtime. |+| [OpenHands/OpenHands#17390](https://github.com/OpenHands/OpenHands/issues/17390) | mapped | [F25.cloud-sandbox-states](../feature-map/F25-backends-and-cloud.md) | Affected authenticated Cloud account/session and token-provisioning reproduction are unavailable. |+| [OpenHands/OpenHands#17386](https://github.com/OpenHands/OpenHands/issues/17386) | mapped | [F04.grouped-folders](../feature-map/F04-conversation-list.md) | Fresh browser puts existing workspace-bound conversations into No workspace. |+| [OpenHands/OpenHands#17384](https://github.com/OpenHands/OpenHands/issues/17384) | mapped | [F27.launch-child](../feature-map/F27-workspace-tools.md) | A real encrypted provider connection/profile, reviewed file-based worker fixture and authorized model key are unavailable. |+| [OpenHands/OpenHands#17383](https://github.com/OpenHands/OpenHands/issues/17383) | mapped | [F27.browser-screenshot](../feature-map/F27-workspace-tools.md) | Actual pinned runtime browser-use 0.11.13/mcp 1.30.0/fastmcp 3.4.8 BrowserToolExecutor constructs. The report’s supported Canvas/global browser operation still needs a genuine browser-enabled agent launch and authorized LLM to issue a browser_* operation; a constructor probe alone does not establish that user flow. |+| [OpenHands/OpenHands#17371](https://github.com/OpenHands/OpenHands/issues/17371) | mapped | [F25.add-agent-server / F25.recovery-gate](../feature-map/F25-backends-and-cloud.md) | Added QA_Rotation_Legacy for the identical origin through Add Backend using the owned session-key file. After restart --rotate-key and reload, read-only comparison shows default-local matches injected key but the manually registered active entry does not. UI shows Disconnected and Unauthorized; core unreconciled credential persists, although there is visible status feedback in this version. |+| [OpenHands/OpenHands#17271](https://github.com/OpenHands/OpenHands/issues/17271) | mapped | [F05.context-window-meter](../feature-map/F05-composer.md) | A completed conversation with nonzero token usage before New Chat; a dummy-key run supplies no genuine context usage. |+| [OpenHands/OpenHands#17263](https://github.com/OpenHands/OpenHands/issues/17263) | mapped | [F05.context-window-meter](../feature-map/F05-composer.md) | A real successful compaction whose condenser usage remains larger than the default agent usage. |+| [OpenHands/OpenHands#17261](https://github.com/OpenHands/OpenHands/issues/17261) | mapped | [F10.custom-model / F10.connection-link](../feature-map/F10-llm-profiles.md) | A real Advanced UI Save of the exact unqualified qwen model linked to an OpenAI connection succeeded; both subsequent actual Canvas conversations emitted LLM Provider NOT provided model=qwen before HTTP. The reported Windows installation, unlinked preflight and successful local inference remain untested. |+| [OpenHands/OpenHands#17257](https://github.com/OpenHands/OpenHands/issues/17257) | mapped | [F04.start-task-cards](../feature-map/F04-conversation-list.md) | An authenticated Cloud backend producing an in-progress sandbox/repository start task. |+| [OpenHands/OpenHands#17239](https://github.com/OpenHands/OpenHands/issues/17239) | mapped | [F05.llm-profile-picker](../feature-map/F05-composer.md) | Local Home/conversation profile-picker checks passed, but the reported Cloud and real ACP model-picker branches need an approved Cloud or ACP session with a multi-row model catalogue. |+| [OpenHands/OpenHands#17233](https://github.com/OpenHands/OpenHands/issues/17233) | mapped | [F10.model-picker](../feature-map/F10-llm-profiles.md) | Reproduced the Basic entry dead end described by concrete companion #17577. The authenticated Advanced/save branch was not exercised. |+| [OpenHands/OpenHands#17213](https://github.com/OpenHands/OpenHands/issues/17213) | out-of-scope | — | The report is OpenHands 0.54.0 runtime-image construction through ToolShield/rootless Podman, including its legacy runtime Docker template; those runtime-builder paths and UI are absent from this Agent Canvas frontend/Agent Server launcher checkout. |+| [OpenHands/OpenHands#17198](https://github.com/OpenHands/OpenHands/issues/17198) | mapped | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | Affected Windows desktop installation, project definitions and a deterministic failing command fixture are unavailable. |+| [OpenHands/OpenHands#17195](https://github.com/OpenHands/OpenHands/issues/17195) | mapped | [F10.connection-link / F10.custom-model](../feature-map/F10-llm-profiles.md) | A real Advanced UI profile linked to provider=custom_openai at refused loopback and model=ollama-local/qwen saved successfully, then both actual starts emitted LLM Provider NOT provided before HTTP. The qwen suffix is an owned diagnostic fixture because the report omits its exact exposed model ID; actual 9Router/Ollama inference and that exact ID remain untested. |+| [OpenHands/OpenHands#17163](https://github.com/OpenHands/OpenHands/issues/17163) | mapped | [F25.cloud-sandbox-states](../feature-map/F25-backends-and-cloud.md) | Affected Free Cloud account, model entitlements and organization history are unavailable. |+| [OpenHands/OpenHands#17162](https://github.com/OpenHands/OpenHands/issues/17162) | mapped | [F19.launch-trust](../feature-map/F19-plugins.md) | Crafted evil.example/github.com/...source renders onlyAll-Hands-AI/openhands-plugin and omits evil.example twice; Start remained disabled and no plugin was fetched. |+| [OpenHands/OpenHands#17161](https://github.com/OpenHands/OpenHands/issues/17161) | mapped | [F06.markdown-rendering](../feature-map/F06-agent-activity.md) | A real reasoning, activated-skill, tool-description or error-detail body rendered without includeStandard. The attempted skills dialog exposed public skills only and did not reach the affected chat renderer. |+| [OpenHands/OpenHands#17148](https://github.com/OpenHands/OpenHands/issues/17148) | out-of-scope | — | Windows CI lane naming/test gating; repository validation workflow. |+| [OpenHands/OpenHands#17135](https://github.com/OpenHands/OpenHands/issues/17135) | mapped | [F10.subscription](../feature-map/F10-llm-profiles.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/OpenHands#17128](https://github.com/OpenHands/OpenHands/issues/17128) | not-reproduced | [F21.kebab-menu / F21.phone](https://github.com/OpenHands/OpenHands/issues/17128#issuecomment-6121792158) | The 25th list row is scrolled into view and its action menu is opened at desktop 1440×1000 and phone 390×844. The full 160×236 menu and its 150×36 Delete button fit within both viewports; no option is cut off in this setup. |+| [OpenHands/OpenHands#17121](https://github.com/OpenHands/OpenHands/issues/17121) | mapped | [F23.activity-log / F23.run-status-polling](../feature-map/F23-automation-detail.md) | Actual bounded local script run API current_phase=Starting automation, no structured phase fields; Activity Log reads only Running and contains no phase text. |+| [OpenHands/OpenHands#17112](https://github.com/OpenHands/OpenHands/issues/17112) | mapped | [F26.desktop-main-window / F26.desktop-external-links](../feature-map/F26-runtime-variants.md) | A runnable owned Electron shell and repeatable popup/redirect fixtures are unavailable. |+| [OpenHands/OpenHands#17108](https://github.com/OpenHands/OpenHands/issues/17108) | mapped | [F03.open-repository-dialog](../feature-map/F03-home.md) | An authenticated Cloud Git provider and a real repository search response that remains pending after clearing the input. |+| [OpenHands/OpenHands#17096](https://github.com/OpenHands/OpenHands/issues/17096) | out-of-scope | — | Agent Server version default documentation drift; documentation only. |+| [OpenHands/OpenHands#17094](https://github.com/OpenHands/OpenHands/issues/17094) | out-of-scope | — | Development persistence path documentation drift; documentation only. |+| [OpenHands/OpenHands#17092](https://github.com/OpenHands/OpenHands/issues/17092) | out-of-scope | — | Session API-key filename documentation drift; documentation only. |+| [OpenHands/OpenHands#17069](https://github.com/OpenHands/OpenHands/issues/17069) | out-of-scope | — | .env.sample launcher comments; documentation only, with no runtime change requested. |+| [OpenHands/OpenHands#17061](https://github.com/OpenHands/OpenHands/issues/17061) | mapped | [F07.rename-inline](../feature-map/F07-conversation-page.md) | Whitespace sends PATCH title empty; current server returns422 and restores the prior title. |+| [OpenHands/OpenHands#17051](https://github.com/OpenHands/OpenHands/issues/17051) | mapped | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | Codex ACP can work in New Chat while Issue-to-PR automation fails with missing OpenAI API credentials. Reproduction needs an approved Codex subscription login plus a configured Issue-to-PR automation and its repository credentials; the report does not identify the saved automation/profile settings. |+| [OpenHands/OpenHands#17026](https://github.com/OpenHands/OpenHands/issues/17026) | mapped | [F06.thinking](../feature-map/F06-agent-activity.md) | A reasoning-capable live model emitting nonempty MessageEvent.llm_message.reasoning_content on a text-only turn, followed by reload. |+| [OpenHands/OpenHands#16995](https://github.com/OpenHands/OpenHands/issues/16995) | mapped | [F20.rail-entry](../feature-map/F20-canvas-apps.md) | Installed and enabled reviewed local demo; hover extension sidebar entry reveals no pin control, while hover Automate reveals sidebar-pin-home-toggle-automations. |+| [OpenHands/OpenHands#16988](https://github.com/OpenHands/OpenHands/issues/16988) | out-of-scope | — | Mock-LLM E2E reporter success marker after global timeout; test reporting tooling. |+| [OpenHands/OpenHands#16977](https://github.com/OpenHands/OpenHands/issues/16977) | out-of-scope | — | Mock-LLM E2E reporter counts retry attempts; test reporting tooling. |+| [OpenHands/OpenHands#16954](https://github.com/OpenHands/OpenHands/issues/16954) | mapped | [F17.server-health](../feature-map/F17-mcp-servers.md) | Two SPA switches from a freshly probed Local identity to unprobed QA_SCOPE reused the same Reachable verdict. The identities use ingress/direct URLs of one owned stack; independent credential and delayed-probe branches were not tested. |+| [OpenHands/OpenHands#16951](https://github.com/OpenHands/OpenHands/issues/16951) | mapped | [F03.composer-blocked-without-llm](../feature-map/F03-home.md) | Home readiness follows pinned key presence while implicit launch uses a different active LLM. Both directions observed; actual inverse launch used the keyless active model and omitted named AgentProfile. |+| [OpenHands/OpenHands#16950](https://github.com/OpenHands/OpenHands/issues/16950) | mapped | [F03.selection-after-reload](../feature-map/F03-home.md) | Selected secondary workspace remains visible after SPA switch and is submitted to the first Local backend. |+| [OpenHands/OpenHands#16949](https://github.com/OpenHands/OpenHands/issues/16949) | mapped | [F08.files-refresh](../feature-map/F08-workspace-files-and-changes.md) | An authenticated Cloud repository for the refresh-query mismatch, or real file-editor observations causing an in-place M-to-D Git refetch. |+| [OpenHands/OpenHands#16948](https://github.com/OpenHands/OpenHands/issues/16948) | mapped | [F15.save-dirty-state](../feature-map/F15-agent-behavior-settings.md) | Reproduced twice on a healthy frozen stack: an unsaved Condenser Max Size=111 was replaced by the changed server baseline=300 and Save disabled after a real stale-query background refetch. No UI Save, reset or scope change occurred. Clock/online state and Max Size=240 restored. |+| [OpenHands/OpenHands#16947](https://github.com/OpenHands/OpenHands/issues/16947) | mapped | [F06.load-older-history](../feature-map/F06-agent-activity.md) | A real long conversation whose REST boundary separates one action from its observation; retain the separate companion #17559. |+| [OpenHands/OpenHands#16927](https://github.com/OpenHands/OpenHands/issues/16927) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Killed only the owned required static frontend after a healthy startup. The genuine launcher remains alive and authenticated /api/settings still returns 200; UI / returns 502 Bad Gateway. This exercises the required-child fatal-exit branch using SIGKILL, not an EADDRINUSE race; exact trigger is recorded. Normal stack restored afterward. |+| [OpenHands/OpenHands#16905](https://github.com/OpenHands/OpenHands/issues/16905) | mapped | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | An enabled built-in OpenHands skill can be absent from a new Claude Code ACP session. The enabled/disabled instruction-delivery check needs an approved authenticated Claude Code account, its ACP subprocess and the configured Code Review automation; no provider session is authorized here. |+| [OpenHands/OpenHands#16893](https://github.com/OpenHands/OpenHands/issues/16893) | mapped | [F18.page](../feature-map/F18-skills.md) | Two SPA backend switches made zero skills requests and omitted a newly arranged card. Reload then fetched actual data and showed it. Registry identities share one physical stack, with real fixture data changes between loads. |+| [OpenHands/OpenHands#16877](https://github.com/OpenHands/OpenHands/issues/16877) | mapped | [F07.open-by-url](../feature-map/F07-conversation-page.md) | Real Local idle/null-title check used 30-second polls. The Cloud null-title IDLE/FINISHED branch needs an authenticated Cloud fixture with failed/unsupported autotitle. |+| [OpenHands/OpenHands#16876](https://github.com/OpenHands/OpenHands/issues/16876) | mapped | [F07.history-skeleton](../feature-map/F07-conversation-page.md) | A real planning sub-conversation whose batch lookup fails or yields null, with the main history available. |+| [OpenHands/OpenHands#16874](https://github.com/OpenHands/OpenHands/issues/16874) | mapped | [F25.cloud-sandbox-states](../feature-map/F25-backends-and-cloud.md) | An authenticated Cloud account, paused sandbox and controlled single-request failure are unavailable. |+| [OpenHands/OpenHands#16872](https://github.com/OpenHands/OpenHands/issues/16872) | mapped | [F26.desktop-external-links](../feature-map/F26-runtime-variants.md) | A runnable owned Electron shell and inert OAuth authorization URI fixture are unavailable. |+| [OpenHands/OpenHands#16871](https://github.com/OpenHands/OpenHands/issues/16871) | mapped | [F08.vscode-link](../feature-map/F08-workspace-files-and-changes.md) | An authorized Cloud account/sandbox with a controlled editor exposed_urls value for the current shared-transformer scheme-rejection path. The original card action is absent and the current drawer uses noopener,noreferrer; no exploit or Cloud scheme outcome was reproduced. |+| [OpenHands/OpenHands#16851](https://github.com/OpenHands/OpenHands/issues/16851) | mapped | [F05.profile-identity](../feature-map/F05-composer.md) | The original conversation/model-switch history and usage record showing Opus with a Fable selector; a dummy-key conversation has no usage evidence. |+| [OpenHands/OpenHands#16844](https://github.com/OpenHands/OpenHands/issues/16844) | mapped | [F03.workspace-dropdown-search](../feature-map/F03-home.md) | Within2.7seconds of SPA backend switch, saved workspace picker reuses the prior Local registry instead of the active secondary registry. |+| [OpenHands/OpenHands#16842](https://github.com/OpenHands/OpenHands/issues/16842) | mapped | [F06.reconnect](../feature-map/F06-agent-activity.md) | A transport fixture delaying an old socket close until its replacement is open; ordinary server restart cannot select that ordering. |+| [OpenHands/OpenHands#16839](https://github.com/OpenHands/OpenHands/issues/16839) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Twenty sequential genuine ingress authenticated /api/conversations/search requests over one keep-alive connection, allHTTP 200; the owned ingress emitted MaxListenersExceededWarning for11 timeout listeners. |+| [OpenHands/OpenHands#16835](https://github.com/OpenHands/OpenHands/issues/16835) | mapped | [F05.context-window-meter](../feature-map/F05-composer.md) | A working prompt-caching model/proxy and a real conversation reporting nonzero cache_read_tokens. |+| [OpenHands/OpenHands#16802](https://github.com/OpenHands/OpenHands/issues/16802) | mapped | [F08.files-rich-plain](../feature-map/F08-workspace-files-and-changes.md) | Rich fenced code renders nested dark/gray containers and padding; layout-only inspection. |+| [OpenHands/OpenHands#16801](https://github.com/OpenHands/OpenHands/issues/16801) | mapped | [F03.launch-without-workspace](../feature-map/F03-home.md) | A working Claude Opus agent and the original task/prompt that makes it explore the default workspace; the prompt is absent. |+| [OpenHands/OpenHands#16774](https://github.com/OpenHands/OpenHands/issues/16774) | mapped | [F26.desktop-boot-splash / F26.desktop-main-window](../feature-map/F26-runtime-variants.md) | The reported signed/notarized/quarantined macOS release DMG and an owned installation fixture are unavailable. |+| [OpenHands/OpenHands#16752](https://github.com/OpenHands/OpenHands/issues/16752) | mapped | [F04.card-live-status](../feature-map/F04-conversation-list.md) | An empty IDLE conversation uses a green active dot; no pulse animation was observed. |+| [OpenHands/OpenHands#16750](https://github.com/OpenHands/OpenHands/issues/16750) | mapped | [F04.card](../feature-map/F04-conversation-list.md) | A prompt automation or native delegated task that remains running during conversation-list reload; no usable model is configured here. |+| [OpenHands/OpenHands#16748](https://github.com/OpenHands/OpenHands/issues/16748) | out-of-scope | — | MSW/jsdom teardown unhandled rejection; test runner lifecycle. |+| [OpenHands/OpenHands#16735](https://github.com/OpenHands/OpenHands/issues/16735) | mapped | [F19.launch-unicode](../feature-map/F19-plugins.md) | A real UTF-8 base64 launch path corrupted Chinese/emoji parameter intoLatin1mojibake twice, including reload. This tests decoding; no source-builder function evaluation was counted asUIproof. |+| [OpenHands/OpenHands#16727](https://github.com/OpenHands/OpenHands/issues/16727) | mapped | [F07.open-by-url](../feature-map/F07-conversation-page.md) | A disposable VM interrupted at the event-write boundary or the original missing-event persisted fixture. |+| [OpenHands/OpenHands#16726](https://github.com/OpenHands/OpenHands/issues/16726) | mapped | [F10.model-picker](../feature-map/F10-llm-profiles.md) | Azure Basic had no Base URL input; Advanced had one, and returning to Basic hid it again. No authenticated Azure conversation was attempted. |+| [OpenHands/OpenHands#16724](https://github.com/OpenHands/OpenHands/issues/16724) | out-of-scope | — | Vercel dependency-install Git rewrite configuration; deployment build tooling with no current Canvas runtime reproduction. |+| [OpenHands/OpenHands#16720](https://github.com/OpenHands/OpenHands/issues/16720) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | A bounded reproducible CPU-heavy workload, affected resource envelope and model key are unavailable. |+| [OpenHands/OpenHands#16719](https://github.com/OpenHands/OpenHands/issues/16719) | mapped | [F06.live-activity](../feature-map/F06-agent-activity.md) | The event/status trace and deterministic stalled or errored state from the original report; its trigger is unspecified. |+| [OpenHands/OpenHands#16716](https://github.com/OpenHands/OpenHands/issues/16716) | mapped | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | The original simultaneous-agent/shared-repository state or a deterministic stale-diff trigger; the report only suggests that timing. |+| [OpenHands/OpenHands#16690](https://github.com/OpenHands/OpenHands/issues/16690) | mapped | [F26.docker-image / F26.runtime-services](../feature-map/F26-runtime-variants.md) | Affected Windows-host Docker setup and persisted conversation/workload fixture are unavailable. |+| [OpenHands/OpenHands#16689](https://github.com/OpenHands/OpenHands/issues/16689) | mapped | [F08.files-load-error](../feature-map/F08-workspace-files-and-changes.md) | The original fresh-VM URL, auth/cookie state and exact request sequence; the report says the trigger is unclear. |+| [OpenHands/OpenHands#16686](https://github.com/OpenHands/OpenHands/issues/16686) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | A Nebius credential and its rejected DeepSeek-V3 request, or the original complete emitted error fixture. |+| [OpenHands/OpenHands#16611](https://github.com/OpenHands/OpenHands/issues/16611) | mapped | [F22.setup-plugin-create](../feature-map/F22-automation-creation.md) | Authenticated Cloud account, model key and reviewed no-op plugin fixture are unavailable. |+| [OpenHands/OpenHands#16601](https://github.com/OpenHands/OpenHands/issues/16601) | mapped | [F01.telemetry-consent](../feature-map/F01-first-run-and-sign-in.md) | The multi-day trigger and backend consent response when the modal returns; opted-in localStorage alone does not reproduce it. |+| [OpenHands/OpenHands#16583](https://github.com/OpenHands/OpenHands/issues/16583) | out-of-scope | — | Issue-readiness and PR-description validation of fenced examples; repository tooling. |+| [OpenHands/OpenHands#16582](https://github.com/OpenHands/OpenHands/issues/16582) | mapped | [F06.stop-resume](../feature-map/F06-agent-activity.md) | A maximum-iteration legacy error delivered after socket replacement, with a real model run reaching that limit. |+| [OpenHands/OpenHands#16581](https://github.com/OpenHands/OpenHands/issues/16581) | mapped | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | A supported host fixture with simultaneous independent Terminal consumers and real command histories is unavailable. |+| [OpenHands/OpenHands#16580](https://github.com/OpenHands/OpenHands/issues/16580) | mapped | [F05.composer-resize](../feature-map/F05-composer.md) | Real touch drags or a CLI touch-gesture verb. Current cleanup already removes capture listeners with capture=true; this CLI exposes mouse drag only. |+| [OpenHands/OpenHands#16579](https://github.com/OpenHands/OpenHands/issues/16579) | mapped | [F06.load-older-history](../feature-map/F06-agent-activity.md) | A delayed older-history response released after another conversation loads; current source has an active-load guard, but the exact race needs a transport fixture. |+| [OpenHands/OpenHands#16578](https://github.com/OpenHands/OpenHands/issues/16578) | not-reproduced | [F10.default-used / F10.custom-model](https://github.com/OpenHands/OpenHands/issues/16578#issuecomment-6122466700) | Two actual Canvas Bedrock starts reached boto3.Session and the expected qa_nonexistent profile-not-found failure; their real traces did not contain No module named boto3. AWS metadata was disabled in the owned server environment, and both runtime/base endpoints were refused loopback. This establishes current Local SDK1.54.0 named-profile import, not valid AWS credentials, successful inference, historical Linux dependency resolution or fresh-install variants. |+| [OpenHands/OpenHands#16551](https://github.com/OpenHands/OpenHands/issues/16551) | mapped | [F27.browser-screenshot](../feature-map/F27-workspace-tools.md) | Native Windows with both system Chrome and Playwright Chromium plus a model key is unavailable. |+| [OpenHands/OpenHands#16530](https://github.com/OpenHands/OpenHands/issues/16530) | out-of-scope | — | This is a request to disclose an unspecified security report and request a CVE; it names no Canvas behavior, affected feature, vulnerability or reproducible failure, so there is no feature-map surface to classify. |+| [OpenHands/OpenHands#16526](https://github.com/OpenHands/OpenHands/issues/16526) | mapped | [F01.onboarding-choose-agent](../feature-map/F01-first-run-and-sign-in.md) | An authenticated hosted Cloud account with the reported organization-membership mismatch during onboarding. |+| [OpenHands/OpenHands#16511](https://github.com/OpenHands/OpenHands/issues/16511) | mapped | [F26.docker-image](../feature-map/F26-runtime-variants.md) | An owned Docker daemon/image and disposable mount ownership fixture are unavailable. |+| [OpenHands/OpenHands#16506](https://github.com/OpenHands/OpenHands/issues/16506) | out-of-scope | — | MSW/jsdom ProgressEvent test teardown race; test runner lifecycle. |+| [OpenHands/OpenHands#16492](https://github.com/OpenHands/OpenHands/issues/16492) | mapped | [F25.manage-backends / F25.add-agent-server](../feature-map/F25-backends-and-cloud.md) | Read-only same-origin browser inspection records the number of registered backend entries and nonempty plaintext apiKey values; the retained probe now shows both owned entries are readable. No credential value or whole storage JSON was printed. |+| [OpenHands/OpenHands#16455](https://github.com/OpenHands/OpenHands/issues/16455) | mapped | [F06.events-match](../feature-map/F06-agent-activity.md) | The original duplicated event/stream trace or deterministic prompt and timing; the report provides no reproducible trigger. |+| [OpenHands/OpenHands#16454](https://github.com/OpenHands/OpenHands/issues/16454) | mapped | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | The original Git changes response and update sequence producing duplicate rows; the report provides no trigger. |+| [OpenHands/OpenHands#16430](https://github.com/OpenHands/OpenHands/issues/16430) | mapped | [F05.attach-files](../feature-map/F05-composer.md) | A remote or throttled real upload that stays in flight long enough to inspect progress and composer state. |+| [OpenHands/OpenHands#16429](https://github.com/OpenHands/OpenHands/issues/16429) | out-of-scope | — | Release-ready workflow permissions; repository CI tooling. |+| [OpenHands/OpenHands#16399](https://github.com/OpenHands/OpenHands/issues/16399) | mapped | [F04.pin](../feature-map/F04-conversation-list.md) | A page-two pin disappears after first-page reload although the conversation still exists. |+| [OpenHands/OpenHands#16383](https://github.com/OpenHands/OpenHands/issues/16383) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Inspect only the owned static-server process for frontend18813: session-key equals a literal argv value. Evidence records booleans only; no credential/argv string exposed. |+| [OpenHands/OpenHands#16374](https://github.com/OpenHands/OpenHands/issues/16374) | out-of-scope | — | Cloud app-server SQLAlchemy read-session atomicity has no Canvas surface or client-owned contract; implementation is outside this frontend. |+| [OpenHands/OpenHands#16360](https://github.com/OpenHands/OpenHands/issues/16360) | out-of-scope | — | The issue is labeled Legacy Cloud and describes the removed v0 Cloud composer/status layout, not the current Canvas composer. |+| [OpenHands/OpenHands#16356](https://github.com/OpenHands/OpenHands/issues/16356) | mapped | [F07.open-by-url](../feature-map/F07-conversation-page.md) | A working provider key that can be rotated/revoked, an existing conversation using it, and a replacement profile. |+| [OpenHands/OpenHands#16341](https://github.com/OpenHands/OpenHands/issues/16341) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model with native code-explorer delegation and the original repository task; no delegated-tool output can be generated without credentials. |+| [OpenHands/OpenHands#16339](https://github.com/OpenHands/OpenHands/issues/16339) | mapped | [F05.attach-files](../feature-map/F05-composer.md) | An authenticated Cloud upload sandbox and working model to inspect the delivered filename/path notification and read the uploaded file. |+| [OpenHands/OpenHands#16331](https://github.com/OpenHands/OpenHands/issues/16331) | mapped | [F08.files-tree](../feature-map/F08-workspace-files-and-changes.md) | A real streaming/tool-running conversation while selecting a second workspace file. |+| [OpenHands/OpenHands#16300](https://github.com/OpenHands/OpenHands/issues/16300) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | An uncached isolated reproduction of the reported LiteLLM/Rust dependency toolchain is unavailable. |+| [OpenHands/OpenHands#16298](https://github.com/OpenHands/OpenHands/issues/16298) | mapped | [F18.skill-in-chat](../feature-map/F18-skills.md) | A Slack-monitor setup can leave the agent without the expected skill. The report omits the concrete monitor setup and loaded skill name/version; reproducing the runtime also needs an approved model and a configured Slack integration. Listing the bundled catalog alone cannot settle it. |+| [OpenHands/OpenHands#16297](https://github.com/OpenHands/OpenHands/issues/16297) | mapped | [F25.selector-dropdown](../feature-map/F25-backends-and-cloud.md) | Served the unchanged UI on an owned protected non-loopback LAN origin, with auth-required and no injected key. Onboarding default Host Name is Local despite the LAN URL; after authenticating with only the private run key and skipping remaining setup, the actual selector reads Connected Local. This verifies the non-loopback naming branch, not a VM installation. |+| [OpenHands/OpenHands#16296](https://github.com/OpenHands/OpenHands/issues/16296) | mapped | [F13.acp-credentials](../feature-map/F13-agent-profiles.md) | Claude Code credential fields remain visible below the already-connected banner, leading users to think keys are required. The connected-state check needs an approved existing Claude CLI login on the agent-server host; a dummy credential does not prove that signed-in state. |+| [OpenHands/OpenHands#16277](https://github.com/OpenHands/OpenHands/issues/16277) | mapped | [F22.template-card / F22.templates-page](../feature-map/F22-automation-creation.md) | Frozen-source isolated copy with only private @openhands/extensions custom-automation manifest.requires removed; build served correctly from build/ against the healthy owned backend. Actual /automations/templates renders Application Error with TypeError: Cannot read properties of undefined (reading integrations), crashing the entire catalog instead of isolating the malformed entry. Shared source/build untouched. |+| [OpenHands/OpenHands#16263](https://github.com/OpenHands/OpenHands/issues/16263) | mapped | [F05.profile-identity](../feature-map/F05-composer.md) | An errored conversation with an unmatched model shows Select a model despite loaded profiles. |+| [OpenHands/OpenHands#16257](https://github.com/OpenHands/OpenHands/issues/16257) | out-of-scope | — | The issue explicitly scopes an import-time signal-handler ownership/dev-tooling refactor and states there is no user-facing feature or child-service behavior change. No observed Canvas startup/shutdown failure is supplied for the map; concrete required-service failures remain mapped separately in #16927/#18299. |+| [OpenHands/OpenHands#16226](https://github.com/OpenHands/OpenHands/issues/16226) | mapped | [F26.cli-port-in-use](../feature-map/F26-runtime-variants.md) | Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. The explicit ingress override does not choose another frontend port. |+| [OpenHands/OpenHands#16217](https://github.com/OpenHands/OpenHands/issues/16217) | not-reproduced | [F26.runtime-services](https://github.com/OpenHands/OpenHands/issues/16217#issuecomment-6121792551) | Installed real uvicorn 0.37.0 alone in an owned Python 3.12 venv, prepended that bin to the private launcher PATH, then restarted the genuine current launcher. Automation 1.19.3 starts and /api/automation/health returns HTTP 200; the reported external-uvicorn ModuleNotFoundError does not occur in this exact resolver/PATH setup. Original PATH restored and doctor run afterward. |+| [OpenHands/OpenHands#16210](https://github.com/OpenHands/OpenHands/issues/16210) | mapped | [F03.plugin-picker](../feature-map/F03-home.md) | An authenticated Cloud organization with configured account plugins before opening New Chat > Plugins. |+| [OpenHands/OpenHands#16208](https://github.com/OpenHands/OpenHands/issues/16208) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Fedora Rawhide/Python 3.14.6 environment, reported SDK/runtime and model key are unavailable. |+| [OpenHands/OpenHands#16161](https://github.com/OpenHands/OpenHands/issues/16161) | out-of-scope | — | CI/dependency publication stability audit; contributor build and release tooling. |+| [OpenHands/OpenHands#16155](https://github.com/OpenHands/OpenHands/issues/16155) | not-reproduced | [F21.list-row / F21.run-now](https://github.com/OpenHands/OpenHands/issues/16155#issuecomment-6121792984) | The exact reported long title is arranged in a real inactive list row and scrolled into view. Run now is an icon-only 32×32 px control, with no wrapping Run/now text, in the current list layout. |+| [OpenHands/OpenHands#15421](https://github.com/OpenHands/OpenHands/issues/15421) | mapped | [F17.agent-uses-server](../feature-map/F17-mcp-servers.md) | A configured Linear MCP can be ignored while an explicitly disabled Linear skill is used instead. The new-chat and automation paths need a working authorized Linear MCP connection, a concrete issue ID and an approved model; the report's automation ZIP is also needed to replay its exact automation settings. |+| [OpenHands/OpenHands#15430](https://github.com/OpenHands/OpenHands/issues/15430) | mapped | [F17.install-oauth](../feature-map/F17-mcp-servers.md) | An OAuth MCP callback in Docker can redirect to an unexposed dynamic localhost port. Completing the callback needs a Docker stack and an approved OAuth MCP provider/account (the report uses GitLab); a local unauthenticated Test connection cannot reach the redirect. |+| [OpenHands/OpenHands#15433](https://github.com/OpenHands/OpenHands/issues/15433) | mapped | [F06.user-and-agent-messages](../feature-map/F06-agent-activity.md) | A working streaming model and follow-up sent while its response is still arriving. |+| [OpenHands/OpenHands#15434](https://github.com/OpenHands/OpenHands/issues/15434) | mapped | [F26.remote-backend](../feature-map/F26-runtime-variants.md) | Affected non-localhost HTTPS reverse proxy deployment and functioning editor binary are unavailable. |+| [OpenHands/OpenHands#15458](https://github.com/OpenHands/OpenHands/issues/15458) | mapped | [F18.page / F18.project-skill](../feature-map/F18-skills.md) | The global Skills page can list AGENTS.md files from cloned automation-run worktrees. The report says the discovery trigger is unknown; replay needs the affected automation workspace/discovery-root layout or captured request parameters, rather than only a normal project-skill fixture. |+| [OpenHands/OpenHands#15465](https://github.com/OpenHands/OpenHands/issues/15465) | mapped | [F23.configuration-event / F23.activity-log](../feature-map/F23-automation-detail.md) | Externally reachable responder, authorized webhook/release repository and controlled real event delivery are unavailable. |+| [OpenHands/OpenHands#15495](https://github.com/OpenHands/OpenHands/issues/15495) | out-of-scope | — | The request is for missing AWS Bedrock setup documentation at docs.openhands.dev, with no concrete Canvas control or failing UI behavior. F10 consumes profile forms, not the external documentation authoring surface. |+| [OpenHands/OpenHands#15501](https://github.com/OpenHands/OpenHands/issues/15501) | mapped | [F08.files-rich-plain](../feature-map/F08-workspace-files-and-changes.md) | The original preview panel and file URL or minimal page identifying the intended surface; Files Rich deliberately blocks scripts, while Browser is separate. |+| [OpenHands/OpenHands#15509](https://github.com/OpenHands/OpenHands/issues/15509) | mapped | [F22.setup-tarball-create](../feature-map/F22-automation-creation.md) | Native Windows stack and controlled event/webhook delivery fixture are unavailable. |+| [OpenHands/OpenHands#15519](https://github.com/OpenHands/OpenHands/issues/15519) | mapped | [F08.files-load-error](../feature-map/F08-workspace-files-and-changes.md) | The reported Linux Docker deployment/auth topology; this isolated run uses the macOS source launcher. |+| [OpenHands/OpenHands#15527](https://github.com/OpenHands/OpenHands/issues/15527) | mapped | [F06.reconnect](../feature-map/F06-agent-activity.md) | The original custom-domain deployment, idle duration and tab-suspension/network transition; those details are missing. |+| [OpenHands/OpenHands#15543](https://github.com/OpenHands/OpenHands/issues/15543) | mapped | [F07.git-control-bar](../feature-map/F07-conversation-page.md) | The original Git probe stdout/interleaving that temporarily puts a filename in the branch badge; dirty state alone is not a deterministic trigger. |+| [OpenHands/OpenHands#15567](https://github.com/OpenHands/OpenHands/issues/15567) | mapped | [F08.files-load-error](../feature-map/F08-workspace-files-and-changes.md) | A plain-HTTP non-loopback deployment and browser accessing that origin, with workspace-cookie observation. |+| [OpenHands/OpenHands#15599](https://github.com/OpenHands/OpenHands/issues/15599) | out-of-scope | — | The report concerns an agent choosing a Git ref strategy from a free-form task; it identifies no Canvas state, request or Git-control defect. |+| [OpenHands/OpenHands#15606](https://github.com/OpenHands/OpenHands/issues/15606) | mapped | [F06.reload-mid-run](../feature-map/F06-agent-activity.md) | An authenticated Codex ACP session emitting intermediate messages while another window reloads history. |+| [OpenHands/OpenHands#15626](https://github.com/OpenHands/OpenHands/issues/15626) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | The original LiteLLM proxy/database-failure event or a disposable proxy with its PostgreSQL dependency unavailable. |+| [OpenHands/OpenHands#15662](https://github.com/OpenHands/OpenHands/issues/15662) | mapped | [F02.sidebar-collapse](../feature-map/F02-app-shell.md) | Collapsed rail has no existing-conversation chooser; expansion is required. |+| [OpenHands/OpenHands#15676](https://github.com/OpenHands/OpenHands/issues/15676) | mapped | [F25.cloud-sandbox-states](../feature-map/F25-backends-and-cloud.md) | Affected Vercel Cloud session, aged sandbox and exact error/deployment revision are unavailable. |+| [OpenHands/OpenHands#15692](https://github.com/OpenHands/OpenHands/issues/15692) | mapped | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | A real agent commit/push terminal observation while the diff remains mounted, plus a disposable Git remote. |+| [OpenHands/OpenHands#15694](https://github.com/OpenHands/OpenHands/issues/15694) | not-reproduced | [F18.toggle](https://github.com/OpenHands/OpenHands/issues/15694#issuecomment-6121789433) | The disabled skill stayed off in two eight-second samples and reloads; the server retained its disabled name. The report is intermittent, so longer or different timing remains unverified. |+| [OpenHands/OpenHands#15701](https://github.com/OpenHands/OpenHands/issues/15701) | mapped | [F03.launch-in-workspace](../feature-map/F03-home.md) | The original stuck creation request/outcome sequence; the report is intermittent and gives no deterministic trigger. |+| [OpenHands/OpenHands#15778](https://github.com/OpenHands/OpenHands/issues/15778) | mapped | [F10.create-validation](../feature-map/F10-llm-profiles.md) | Fast refused-loopback validation rejected the profile, but two validations with five retries reached the 30-second browser timeout and then saved with 201. |+| [OpenHands/OpenHands#15782](https://github.com/OpenHands/OpenHands/issues/15782) | mapped | [F05.draft-persistence](../feature-map/F05-composer.md) | Clarify whether prompt means unsent composer text or the already-sent history message, and capture that state before sidebar Stop. |+| [OpenHands/OpenHands#15813](https://github.com/OpenHands/OpenHands/issues/15813) | mapped | [F15.save-errors](../feature-map/F15-agent-behavior-settings.md) | Max Size accepted -5 and enabled Save twice. The server rejected the save and kept 240; the client input floor is missing. |+| [OpenHands/OpenHands#15811](https://github.com/OpenHands/OpenHands/issues/15811) | mapped | [F15.verification-page / F15.verification-critic-fields](../feature-map/F15-agent-behavior-settings.md) | Default Basic and Advanced both show two switches. Enabling Confirmation Mode reveals Security Analyzer only on Advanced. The report asks whether this is intentional; mapped as conditional-tier clarification. |+| [OpenHands/OpenHands#15822](https://github.com/OpenHands/OpenHands/issues/15822) | mapped | [F25.switch-backend / F25.health-status / F25.cloud-org-rows](../feature-map/F25-backends-and-cloud.md) | A real registered authenticated Cloud backend/account/organization is unavailable. |+| [OpenHands/OpenHands#15833](https://github.com/OpenHands/OpenHands/issues/15833) | not-reproduced | [F21.add-menu / F22.agent-assisted-setup](https://github.com/OpenHands/OpenHands/issues/15833#issuecomment-6121793408) | Add Automation → Create Automation opens /automations/setup with a New Automation draft editor, prompt, schedule and assistance input. The old modal-to-empty-conversation branch is replaced on this target, so its missing prefilled chat text does not reproduce. |+| [OpenHands/OpenHands#15946](https://github.com/OpenHands/OpenHands/issues/15946) | mapped | [F04.card](../feature-map/F04-conversation-list.md) | A copied lease-bearing persisted conversation and second owner/expired-lease state; never alter another live run's state. |+| [OpenHands/software-agent-sdk#5718](https://github.com/OpenHands/software-agent-sdk/issues/5718) | mapped | [F19.enabled-autoload](../feature-map/F19-plugins.md) | Two malformed-manifest first sends showed Invalid JSON and Sending, persisted no events and stayed idle. A malformed skill under a valid manifest instead initialized successfully. |+| [OpenHands/software-agent-sdk#5711](https://github.com/OpenHands/software-agent-sdk/issues/5711) | out-of-scope | — | Canvas does not use TypeScript RemoteState; it owns event/state hydration through its frontend stores. |+| [OpenHands/software-agent-sdk#5652](https://github.com/OpenHands/software-agent-sdk/issues/5652) | out-of-scope | — | Python RemoteConversation cached getter behavior; Canvas consumes typed HTTP/WebSocket payloads rather than Python RemoteState. |+| [OpenHands/software-agent-sdk#5651](https://github.com/OpenHands/software-agent-sdk/issues/5651) | out-of-scope | — | Node 22 ConversationEventStream reconnect path; Canvas uses browser sockets and requires Node >=24 for its launcher. |+| [OpenHands/software-agent-sdk#5650](https://github.com/OpenHands/software-agent-sdk/issues/5650) | out-of-scope | — | RemoteWorkspace.downloadAsBlob/fileDownload are not Canvas download consumers; Canvas uses binary-safe FileClient.downloadFile. |+| [OpenHands/software-agent-sdk#5649](https://github.com/OpenHands/software-agent-sdk/issues/5649) | out-of-scope | — | Canvas builds its create payload and calls ConversationClient directly, not TypeScript ConversationManager.createConversation. |+| [OpenHands/software-agent-sdk#5648](https://github.com/OpenHands/software-agent-sdk/issues/5648) | mapped | [F20.app-backend-view](../feature-map/F20-canvas-apps.md) | Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas. |+| [OpenHands/software-agent-sdk#5647](https://github.com/OpenHands/software-agent-sdk/issues/5647) | out-of-scope | — | OpenAI-compatible gateway per-turn usage contract; Canvas does not consume /v1/chat/completions gateway usage. |+| [OpenHands/software-agent-sdk#5646](https://github.com/OpenHands/software-agent-sdk/issues/5646) | mapped | [F20.app-backend-view](../feature-map/F20-canvas-apps.md) | Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas. |+| [OpenHands/software-agent-sdk#5645](https://github.com/OpenHands/software-agent-sdk/issues/5645) | mapped | [F19.add-plugin](../feature-map/F19-plugins.md) | Two Add attempts with an unknown ref on a warm local Git cache succeeded and installed cached version 1.0.0 instead of rejecting the missing ref. |+| [OpenHands/software-agent-sdk#5644](https://github.com/OpenHands/software-agent-sdk/issues/5644) | out-of-scope | — | Canvas's LlmSettingsLocalView.handleSave explicitly skips ProfilesService.validateProfile when connectionId is set. The linked-draft validation request in this SDK report is not emitted by the mapped Canvas profile-save path. |+| [OpenHands/software-agent-sdk#5643](https://github.com/OpenHands/software-agent-sdk/issues/5643) | mapped | [F17.custom-add / F17.agent-uses-server](../feature-map/F17-mcp-servers.md) | An empty MCP body returned 201 twice and persisted enabled:true; real UI-started initialization lost the healthy echo tool. The invalid entry had no UI card. |+| [OpenHands/software-agent-sdk#5642](https://github.com/OpenHands/software-agent-sdk/issues/5642) | mapped | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | Tracked my notes.md modification makes scoped git/changes return400 and hides changes. |+| [OpenHands/software-agent-sdk#5641](https://github.com/OpenHands/software-agent-sdk/issues/5641) | mapped | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | Plain parent workspace shows No commits yet while its child repo has modified/untracked files; removed only run-owned auto-created empty .git to arrange the reported layout. |+| [OpenHands/software-agent-sdk#5640](https://github.com/OpenHands/software-agent-sdk/issues/5640) | mapped | [F07.download-zip](../feature-map/F07-conversation-page.md) | A deferred-init warm-pool Agent Server initialized with a non-default conversations_path; this launcher starts an already-initialized server. |+| [OpenHands/software-agent-sdk#5639](https://github.com/OpenHands/software-agent-sdk/issues/5639) | mapped | [F05.llm-profile-picker](../feature-map/F05-composer.md) | A working model, a profile switch before the first run, and nonzero usage/cost after that turn. |+| [OpenHands/software-agent-sdk#5638](https://github.com/OpenHands/software-agent-sdk/issues/5638) | mapped | [F08.files-tree](../feature-map/F08-workspace-files-and-changes.md) | A real1800-file listing over1MiB drops README.md and creates a partial-path bb folder before the2000-file cap. |+| [OpenHands/software-agent-sdk#5637](https://github.com/OpenHands/software-agent-sdk/issues/5637) | mapped | [F20.app-backend-view](../feature-map/F20-canvas-apps.md) | Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas. |+| [OpenHands/software-agent-sdk#5636](https://github.com/OpenHands/software-agent-sdk/issues/5636) | out-of-scope | — | Canvas direct profile switches assign a new profile:<name>:<uuid> usage_id; the raw switch_llm default/already-registered usage_id collision is not its request shape. |+| [OpenHands/software-agent-sdk#5635](https://github.com/OpenHands/software-agent-sdk/issues/5635) | mapped | [F06.reconnect](../feature-map/F06-agent-activity.md) | prepare-for-sandbox-pause returned204 but the original communicating session socket stayed open for over10s. A newly opened Canvas tab received a new run:false message; the old tab did not. DELETE returned200 and GET404, but both held sockets stayed open over10s. Canvas later redirected through metadata polling; no claim of permanently stale navigation is made. |+| [OpenHands/software-agent-sdk#5633](https://github.com/OpenHands/software-agent-sdk/issues/5633) | mapped | [F26.remote-backend](../feature-map/F26-runtime-variants.md) | Temporarily replaced only the owned Agent Server with the same pinned interpreter and state under genuine default Rich logging (LOG_JSON/CI/LOG_RICH_TRACEBACKS unset). Canvas Send uses tools=[] and a dummy LLM pointing at closed 127.0.0.1:9; 60 concurrent direct /alive probes had maximum latency 2.091 s while the conversation ended error. No provider reachable. Restored normal launcher and doctor afterward. |+| [OpenHands/software-agent-sdk#5632](https://github.com/OpenHands/software-agent-sdk/issues/5632) | mapped | [F11.list](../feature-map/F11-provider-connections.md) | The own provider-store lock delayed unrelated /alive responses by 25.416 and 28.060 seconds; the second list request returned 503 after 30.052 seconds. Locks were released and doctor passed. |+| [OpenHands/software-agent-sdk#5631](https://github.com/OpenHands/software-agent-sdk/issues/5631) | out-of-scope | — | SDK event-search short-kind alias contract; current Canvas event history requests do not send that short kind filter. |+| [OpenHands/software-agent-sdk#5630](https://github.com/OpenHands/software-agent-sdk/issues/5630) | mapped | [F03.launch-in-workspace / F03.create-error-toast](../feature-map/F03-home.md) | Owned workspace has .agents/agents/qa-bad-skill.md naming nonexistent qa-f27-missing-skill. With only a closed-loopback dummy LLM profile and autotitling disabled, Home launch in that folder returns the missing-skill error and remains on Home. Two genuine message POSTs with run:false and a create with initial_message run:false each returned HTTP 500 before inference. |+| [OpenHands/software-agent-sdk#5629](https://github.com/OpenHands/software-agent-sdk/issues/5629) | mapped | [F07.rename-inline](../feature-map/F07-conversation-page.md) | B was started with shared storage before A renamed an idle conversation from BEFORE_B to AFTER_B. A PATCH and disk confirmed AFTER_B. After own A graceful stop, native B events/count returned200 and rewrote disk title to its BEFORE_B startup copy. Real Add Backend→B card showed BEFORE_B in Canvas. The native first-load request is explicit; an earlier sandbox-pause variant did not reach a valid first load and is not the claimed reproduction. |+| [OpenHands/software-agent-sdk#5628](https://github.com/OpenHands/software-agent-sdk/issues/5628) | mapped | [F19.enabled-autoload](../feature-map/F19-plugins.md) | Two actual Local conversations initialized an enabled installed plugin, then resumed after real UI disable plus server restart. Each resumed command still activated qa-resume-plugin:qa-hello and include_skills retained both plugin skills, while the initial hook count stayed one. A fresh chat begun while disabled had no command activation or hook. All messages reached only refused loopback model initialization; command activation and retained state are pre-model observables, not successful command instructions or inference. |+| [OpenHands/software-agent-sdk#5627](https://github.com/OpenHands/software-agent-sdk/issues/5627) | mapped | [F05.slash-goal](../feature-map/F05-composer.md) | A working model goal loop active during graceful backend restart, then goal-status and resume readback. |+| [OpenHands/software-agent-sdk#5626](https://github.com/OpenHands/software-agent-sdk/issues/5626) | not-reproduced | [F06.stop-resume](https://github.com/OpenHands/software-agent-sdk/issues/5626#issuecomment-6122465657) | Actual Canvas Stop displayed Stopped; GET and base_state.json both immediately reported paused while a dummy loopback model request was held. The exact launcher-owned Agent Server PID was SIGKILLed, then the own run restarted; GET, disk and reloaded UI retained paused. The accept-only fixture received one connection, generated no response, and bounded-exited cleanly. |+| [OpenHands/software-agent-sdk#5624](https://github.com/OpenHands/software-agent-sdk/issues/5624) | mapped | [F11.create-validation / F11.edit](../feature-map/F11-provider-connections.md) | Two real Add forms accepted the redacted placeholder and persisted keyless connections. Typing it in Edit changed key-set to true. Keyboard Save was used; whitespace-only and provider-auth branches were not claimed. |+| [OpenHands/software-agent-sdk#5622](https://github.com/OpenHands/software-agent-sdk/issues/5622) | mapped | [F14.create / F14.edit / F14.delete](../feature-map/F14-secrets.md) | With an approved replacement-cipher/old-encrypted-state fixture, secret creation rewrote an old entry and its dummy getter returned 404. Two unrelated settings saves cleared the old LLM key. Original cipher and stores were restored immediately; restart and doctor passed. |+| [OpenHands/software-agent-sdk#5607](https://github.com/OpenHands/software-agent-sdk/issues/5607) | mapped | [F17.credential-probe](../feature-map/F17-mcp-servers.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/software-agent-sdk#5590](https://github.com/OpenHands/software-agent-sdk/issues/5590) | out-of-scope | — | Python create_app(Config) lifespan embedding contract; Canvas launches agent-server using its CLI/environment configuration. |+| [OpenHands/software-agent-sdk#5581](https://github.com/OpenHands/software-agent-sdk/issues/5581) | mapped | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | Two owned Agent Server runtimes sharing a disposable TMUX_TMPDIR and a model key for the mapped Terminal path are unavailable. |+| [OpenHands/software-agent-sdk#5539](https://github.com/OpenHands/software-agent-sdk/issues/5539) | mapped | [F07.download-zip](../feature-map/F07-conversation-page.md) | A20-second three-client real loopback overlapping receiver made929 attempts:31 completed corrupt200 ZIPs,106HTTP500,269ingress502,133RemoteProtocolError,388 validZIPs and2 intentionally deadline-aborted reads. Both ambiguous Content-Length differences were excluded from the public claim. Actual Canvas Download during overlap produced a ZIP with inspectable entries; that particular UI download is not claimed corrupt. A separate sequential2000-request pass returned validZIPs throughout. |+| [OpenHands/software-agent-sdk#5533](https://github.com/OpenHands/software-agent-sdk/issues/5533) | mapped | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | A model key for real agent-driven first-command observations is unavailable. |+| [OpenHands/software-agent-sdk#5523](https://github.com/OpenHands/software-agent-sdk/issues/5523) | mapped | [F05.image-only-send](../feature-map/F05-composer.md) | A working DeepSeek V4.1 Flash vision profile and another fallback profile, with an image-bearing turn. |+| [OpenHands/software-agent-sdk#5519](https://github.com/OpenHands/software-agent-sdk/issues/5519) | mapped | [F13.openhands-options / F13.active-drives-conversation](../feature-map/F13-agent-profiles.md) | A complete scoped worker turn needs an approved working model and the affected legacy profile to dispatch a child and invoke TaskTrackingTool. Current Canvas Save added task_tracker despite its unchecked box, and real parent initialization exposed task/task_tracker; this is not a child-turn verdict. |+| [OpenHands/software-agent-sdk#5501](https://github.com/OpenHands/software-agent-sdk/issues/5501) | mapped | [F20.rail-label](../feature-map/F20-canvas-apps.md) | Existing Agent Server nav_label omission note and F20.rail-label recipe preserved. |+| [OpenHands/software-agent-sdk#5498](https://github.com/OpenHands/software-agent-sdk/issues/5498) | mapped | [F11.delete-stale-reference](../feature-map/F11-provider-connections.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/software-agent-sdk#5497](https://github.com/OpenHands/software-agent-sdk/issues/5497) | mapped | [F11.rotate](../feature-map/F11-provider-connections.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/software-agent-sdk#5496](https://github.com/OpenHands/software-agent-sdk/issues/5496) | mapped | [F19.refresh / F19.start-conversation / F19.enabled-autoload](../feature-map/F19-plugins.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/software-agent-sdk#5495](https://github.com/OpenHands/software-agent-sdk/issues/5495) | mapped | [F15.condenser-kind](../feature-map/F15-agent-behavior-settings.md) | Already linked by this feature map; preserving the recorded observation and dates without claiming a new live run. |+| [OpenHands/software-agent-sdk#5492](https://github.com/OpenHands/software-agent-sdk/issues/5492) | mapped | [F05.slash-goal](../feature-map/F05-composer.md) | A real goal judge response containing complete as the string false rather than a JSON boolean. |+| [OpenHands/software-agent-sdk#5481](https://github.com/OpenHands/software-agent-sdk/issues/5481) | out-of-scope | — | Empty bug template identifies no behavior, reproduction, version or mapped Canvas feature; needs a concrete report before Canvas relevance can be established. |+| [OpenHands/software-agent-sdk#5475](https://github.com/OpenHands/software-agent-sdk/issues/5475) | out-of-scope | — | Enterprise SDK dependency-bump CI permissions; no Canvas feature or user flow. |+| [OpenHands/software-agent-sdk#5466](https://github.com/OpenHands/software-agent-sdk/issues/5466) | mapped | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | Two real Canvas Codex ACP starts resolved acp_command=['codex-acp'] and initialized the run-owned published @zed-industries/[email protected] ahead of PATH despite the SDK registry's 1.10.0 pin. Both actual logs reported agent_version=0.16.0 and the mismatch warning still blamed npx fallback. Initialization then stopped at Authentication required, with private HOME and provider routes refused on loopback. This proves stale binary selection/warning before auth, not a usable login or historical model-cache failure. |+| [OpenHands/software-agent-sdk#5465](https://github.com/OpenHands/software-agent-sdk/issues/5465) | mapped | [F06.branch-from-here](../feature-map/F06-agent-activity.md) | A working cipher-backed model conversation, then fork and a successful turn on the fork. |+| [OpenHands/software-agent-sdk#5447](https://github.com/OpenHands/software-agent-sdk/issues/5447) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | Anthropic Claude Opus5.5 credentials and a real trailing-assistant-prefill request. |+| [OpenHands/software-agent-sdk#5375](https://github.com/OpenHands/software-agent-sdk/issues/5375) | mapped | [F17.install-oauth / F17.agent-uses-server](../feature-map/F17-mcp-servers.md) | A successful GitLab OAuth login can be followed by an invalid localhost redirect URI on the first ticket-to-code-change MCP action. Reproduction needs the authorized self-hosted GitLab MCP endpoint/account and the ticket-to-code-change plugin/workflow; the reported endpoint is not a public fixture. |+| [OpenHands/software-agent-sdk#5373](https://github.com/OpenHands/software-agent-sdk/issues/5373) | mapped | [F26.remote-backend](../feature-map/F26-runtime-variants.md) | A separately hosted default-Rich backend and the original cyclic-exception peer/backlog driver are unavailable. |+| [OpenHands/software-agent-sdk#5354](https://github.com/OpenHands/software-agent-sdk/issues/5354) | mapped | [F06.event-groups](../feature-map/F06-agent-activity.md) | A live model turn mixing valid, invalid and valid tool calls in one response. |+| [OpenHands/software-agent-sdk#5330](https://github.com/OpenHands/software-agent-sdk/issues/5330) | mapped | [F05.slash-btw](../feature-map/F05-composer.md) | A working caching provider and comparable ordinary-message versus /btw prompt/cache records. |+| [OpenHands/software-agent-sdk#5322](https://github.com/OpenHands/software-agent-sdk/issues/5322) | mapped | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | A model key for genuine terminal-tool execution is unavailable; the PowerShell half additionally requires Windows. |+| [OpenHands/software-agent-sdk#5310](https://github.com/OpenHands/software-agent-sdk/issues/5310) | mapped | [F26.docker-conversation-runtime](../feature-map/F26-runtime-variants.md) | Docker daemon/per-conversation image and real auxiliary-title/ACP credentials are unavailable. |+| [OpenHands/software-agent-sdk#5307](https://github.com/OpenHands/software-agent-sdk/issues/5307) | mapped | [F26.docker-conversation-runtime](../feature-map/F26-runtime-variants.md) | Docker daemon/per-conversation image is unavailable; a host/container git fixture is required. |+| [OpenHands/software-agent-sdk#5279](https://github.com/OpenHands/software-agent-sdk/issues/5279) | mapped | [F23.run-linked-conversation](../feature-map/F23-automation-detail.md) | An authenticated Cloud automation workspace/callback context and owned stable conversation are unavailable. |+| [OpenHands/software-agent-sdk#5268](https://github.com/OpenHands/software-agent-sdk/issues/5268) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | A real provider response carrying prompt_tokens_details on the affected LiteLLM/SDK combination. |+| [OpenHands/software-agent-sdk#5252](https://github.com/OpenHands/software-agent-sdk/issues/5252) | out-of-scope | — | Cross-version test-fixture gap; no reported user-visible failure to reproduce. |+| [OpenHands/software-agent-sdk#5225](https://github.com/OpenHands/software-agent-sdk/issues/5225) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | A real provider-rejected poisoned image tool_result in a disposable persisted history, followed by retry. |+| [OpenHands/software-agent-sdk#5213](https://github.com/OpenHands/software-agent-sdk/issues/5213) | mapped | [F05.slash-btw](../feature-map/F05-composer.md) | A working /btw model response with prompt_tokens_details on an affected LiteLLM version. |+| [OpenHands/software-agent-sdk#5207](https://github.com/OpenHands/software-agent-sdk/issues/5207) | mapped | [F11.agent-uses-connection / F11.rotate](../feature-map/F11-provider-connections.md) | OpenRouter linked profiles can lose connection credentials at raw agent/condenser startup or reuse stale credentials after rotation; opaque reasoning can also be lost on persisted turns. End-to-end checks need an approved OpenRouter connection and a model returning reasoning_details, including the linked-profile switch and reload path; loopback failure does not prove them. |+| [OpenHands/software-agent-sdk#5181](https://github.com/OpenHands/software-agent-sdk/issues/5181) | mapped | [F10.schema-validation](../feature-map/F10-llm-profiles.md) | The SDK can turn an advertised output limit into a requested default cap and silently drop an explicitly enabled reasoning_effort. The narrowed maintainer scope needs an authenticated OpenRouter/proxy route exposing lower-limit endpoints and capability metadata, or a real endpoint accepting the explicit override; a synthetic empty-metadata response would not establish production behavior. |+| [OpenHands/software-agent-sdk#5168](https://github.com/OpenHands/software-agent-sdk/issues/5168) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | A working provider response whose usage has no cache-creation field. |+| [OpenHands/software-agent-sdk#5167](https://github.com/OpenHands/software-agent-sdk/issues/5167) | mapped | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | On a headless Docker host an already usable Codex ChatGPT login can still time out during ACP authenticate. Reproduction needs an approved ChatGPT login in the reported headless/read-only-auth-file deployment and its adapter; a desktop or bogus login cannot establish this conditional trigger. |+| [OpenHands/software-agent-sdk#5152](https://github.com/OpenHands/software-agent-sdk/issues/5152) | not-reproduced | [F27.browser-screenshot](https://github.com/OpenHands/software-agent-sdk/issues/5152#issuecomment-6121793764) | The actual unlocked pinned runtime resolves openhands-tools 1.54.0, browser-use 0.11.13, mcp 1.30.0 and fastmcp 3.4.8. Its genuine BrowserToolExecutor constructs successfully using its own interpreter and private HOME. The reported mcp 2.x/list_tools import failure does not occur in this exact dependency set; no model was called. |+| [OpenHands/software-agent-sdk#5138](https://github.com/OpenHands/software-agent-sdk/issues/5138) | out-of-scope | — | Retired TypeScript cloud-proxy body API; current Canvas makes direct Cloud App-API calls and uses typed runtime clients. |+| [OpenHands/software-agent-sdk#5108](https://github.com/OpenHands/software-agent-sdk/issues/5108) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | An Agent Server configured with the native Gemini edit tool set and a real tool edit of an invalid-UTF8 fixture; Gemini ACP uses a different external executor. |+| [OpenHands/software-agent-sdk#5100](https://github.com/OpenHands/software-agent-sdk/issues/5100) | mapped | [F10.create-validation](../feature-map/F10-llm-profiles.md) | The slow branch reproduced twice: five-retry loopback validation lasted until the 30-second client abort, then Canvas saved the profile. The separate exhausted-quota 429 branch requires an approved real endpoint. |+| [OpenHands/software-agent-sdk#5099](https://github.com/OpenHands/software-agent-sdk/issues/5099) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | A working model response with LiteLLM PromptTokensDetailsWrapper lacking cache_creation_tokens. |+| [OpenHands/software-agent-sdk#5096](https://github.com/OpenHands/software-agent-sdk/issues/5096) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model invoking canonical file-editor str_replace/insert through a real symlink, then undo. |+| [OpenHands/software-agent-sdk#5094](https://github.com/OpenHands/software-agent-sdk/issues/5094) | mapped | [F07.open-by-url](../feature-map/F07-conversation-page.md) | An authenticated Codex ACP session whose original session/load fails while persisted history remains. |+| [OpenHands/software-agent-sdk#5092](https://github.com/OpenHands/software-agent-sdk/issues/5092) | mapped | [F06.confirmation-mode](../feature-map/F06-agent-activity.md) | A working model proposing a real confirmation-gated marker action, then a revocation user message. |+| [OpenHands/software-agent-sdk#5074](https://github.com/OpenHands/software-agent-sdk/issues/5074) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A native delegated child still running during an owned server crash and parent recovery. |+| [OpenHands/software-agent-sdk#5072](https://github.com/OpenHands/software-agent-sdk/issues/5072) | out-of-scope | — | TypeScript Babel override/lockfile consistency for builds; no Canvas runtime feature. |+| [OpenHands/software-agent-sdk#5038](https://github.com/OpenHands/software-agent-sdk/issues/5038) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A live canonical file-editor view/edit on the specific UTF8 sample alignment; Canvas Files preview uses a separate NUL sniff, not this validator. |+| [OpenHands/software-agent-sdk#5027](https://github.com/OpenHands/software-agent-sdk/issues/5027) | out-of-scope | — | PR linked-issue readiness-check repository identity; repository CI tooling. |+| [OpenHands/software-agent-sdk#4990](https://github.com/OpenHands/software-agent-sdk/issues/4990) | mapped | [F06.stop-resume](../feature-map/F06-agent-activity.md) | An authenticated Claude Code ACP turn executing separate marker-writing tools while Pause is pressed. |+| [OpenHands/software-agent-sdk#4987](https://github.com/OpenHands/software-agent-sdk/issues/4987) | mapped | [F04.card](../feature-map/F04-conversation-list.md) | A working autotitle model and captured title-generation input/output; Canvas only consumes the resulting server title. |+| [OpenHands/software-agent-sdk#4983](https://github.com/OpenHands/software-agent-sdk/issues/4983) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A runtime without ripgrep and a live agent grep call using extended regex and escaped literal punctuation. |+| [OpenHands/software-agent-sdk#4965](https://github.com/OpenHands/software-agent-sdk/issues/4965) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | A strict OpenAI-compatible provider and a real history containing blank text or empty content arrays. |+| [OpenHands/software-agent-sdk#4963](https://github.com/OpenHands/software-agent-sdk/issues/4963) | mapped | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | Native Windows PowerShell terminal runtime is unavailable. |+| [OpenHands/software-agent-sdk#4934](https://github.com/OpenHands/software-agent-sdk/issues/4934) | mapped | [F10.schema-validation / F10.default-used](../feature-map/F10-llm-profiles.md) | Switching to a MiniMax M3 profile can fail because the SDK sends unsupported reasoning_effort. The runtime switch needs approved working credentials for the starting model and MiniMax route; the reported provider rejection cannot be proved with a loopback connection error. |+| [OpenHands/software-agent-sdk#4912](https://github.com/OpenHands/software-agent-sdk/issues/4912) | out-of-scope | — | SDK CI path-gating and downstream test selection; no Canvas runtime feature. |+| [OpenHands/software-agent-sdk#4910](https://github.com/OpenHands/software-agent-sdk/issues/4910) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | An owned authenticated ACP provider runtime/command chain is unavailable. |+| [OpenHands/software-agent-sdk#4901](https://github.com/OpenHands/software-agent-sdk/issues/4901) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | An owned authenticated ACP provider runtime/command chain is unavailable. |+| [OpenHands/software-agent-sdk#4893](https://github.com/OpenHands/software-agent-sdk/issues/4893) | mapped | [F07.status-menu](../feature-map/F07-conversation-page.md) | A disposable sandbox/VM frozen mid-run with a live model and persisted ownership state. |+| [OpenHands/software-agent-sdk#4891](https://github.com/OpenHands/software-agent-sdk/issues/4891) | not-reproduced | [F17.agent-uses-server](https://github.com/OpenHands/software-agent-sdk/issues/4891#issuecomment-6121790068) | Missing stdio and unreachable HTTP MCP sources both allowed initialization and preserved the healthy sibling tool, then reached the expected loopback model error. No successful reply was claimed. |+| [OpenHands/software-agent-sdk#4887](https://github.com/OpenHands/software-agent-sdk/issues/4887) | out-of-scope | — | SDK release publication workflow dispatch; no Canvas feature. |+| [OpenHands/software-agent-sdk#4847](https://github.com/OpenHands/software-agent-sdk/issues/4847) | out-of-scope | — | Python RemoteConversation.fork() wrapper title parsing; Canvas calls the typed REST fork endpoint directly. |+| [OpenHands/software-agent-sdk#4846](https://github.com/OpenHands/software-agent-sdk/issues/4846) | out-of-scope | — | Node ESM WebSocket constructor selection; Canvas browser selects window.WebSocket. |+| [OpenHands/software-agent-sdk#4811](https://github.com/OpenHands/software-agent-sdk/issues/4811) | out-of-scope | — | The report is about standalone SDK fork title injection into ConversationState.tags; it explicitly separates Agent Server StoredConversation title handling used by Canvas. |+| [OpenHands/software-agent-sdk#4810](https://github.com/OpenHands/software-agent-sdk/issues/4810) | out-of-scope | — | Canvas sets policy/analyzer/secrets at creation or through account settings/resume secrets; it does not call these per-conversation mutation endpoints. |+| [OpenHands/software-agent-sdk#4800](https://github.com/OpenHands/software-agent-sdk/issues/4800) | mapped | [F07.rename-inline](../feature-map/F07-conversation-page.md) | After a successful baseline UI rename, chmod0500 on only the owned conversation directory made the next actual UI rename fail with PermissionError/HTTP500. GET nevertheless reported the rejected title while meta.json retained the baseline. Permissions were restored; graceful restart then persisted the rejected in-memory title. Tags/autotitle branches were not tested. |+| [OpenHands/software-agent-sdk#4798](https://github.com/OpenHands/software-agent-sdk/issues/4798) | out-of-scope | — | TypeScript client endpoint-audit CI comment permissions; no Canvas feature. |+| [OpenHands/software-agent-sdk#4777](https://github.com/OpenHands/software-agent-sdk/issues/4777) | mapped | [F06.stop-resume](../feature-map/F06-agent-activity.md) | A live batch with a second real tool waiting on a declared-resource lock when Stop is pressed. |+| [OpenHands/software-agent-sdk#4776](https://github.com/OpenHands/software-agent-sdk/issues/4776) | out-of-scope | — | There is no structuredContent, structured_content, result_meta or resultMeta consumer under current src/. The report demonstrates SDK conversion/persistence metadata loss, without a current mapped Canvas renderer or API consumer for those fields. |+| [OpenHands/software-agent-sdk#4774](https://github.com/OpenHands/software-agent-sdk/issues/4774) | out-of-scope | — | Python MessageEvent.visualize() CLI terminal rendering; Canvas renders MessageEvent content in its own React components. |+| [OpenHands/software-agent-sdk#4769](https://github.com/OpenHands/software-agent-sdk/issues/4769) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | A working Responses-mode provider whose stream wrapper yields completion but retains completed_response=None. |+| [OpenHands/software-agent-sdk#4765](https://github.com/OpenHands/software-agent-sdk/issues/4765) | out-of-scope | — | The affected redact_text_secrets helper is used by SDK command and ACP diagnostic logging. This report supplies only a direct Python helper call and no affected Canvas-rendered event, endpoint field or user control; F14 secret CRUD does not call it. |+| [OpenHands/software-agent-sdk#4709](https://github.com/OpenHands/software-agent-sdk/issues/4709) | not-reproduced | [F15.agent-context-effect](https://github.com/OpenHands/software-agent-sdk/issues/4709#issuecomment-6121790419) | Canvas launches used fresh runtime datetimes despite naturally stale settings and an explicit old-date fixture. The old date was absent from the initialized prompt. |+| [OpenHands/software-agent-sdk#4705](https://github.com/OpenHands/software-agent-sdk/issues/4705) | mapped | [F23.run-status-polling / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Affected client/server versions, GitHub tool-definition event fixture and a model key for genuine run/status comparison are unavailable. |+| [OpenHands/software-agent-sdk#4670](https://github.com/OpenHands/software-agent-sdk/issues/4670) | out-of-scope | — | Python WebSocketCallbackClient thread shutdown; not the Canvas browser WebSocket path. |+| [OpenHands/software-agent-sdk#4668](https://github.com/OpenHands/software-agent-sdk/issues/4668) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A live planner create/edit yielding PlanningFileEditorObservation with old/new contents for the chat diff. |+| [OpenHands/software-agent-sdk#4667](https://github.com/OpenHands/software-agent-sdk/issues/4667) | out-of-scope | — | Tom conversation index/checkpoint implementation; no mapped Canvas feature or supported Canvas agent path identified. |+| [OpenHands/software-agent-sdk#4666](https://github.com/OpenHands/software-agent-sdk/issues/4666) | out-of-scope | — | The report concerns the backend browser-use rrweb recording controller/flush-task lifecycle. Canvas has no backend recording/replay surface or duplicate-start action; control-openhands browser record is a separate verification-browser implementation and is not this SDK consumer. |+| [OpenHands/software-agent-sdk#4665](https://github.com/OpenHands/software-agent-sdk/issues/4665) | out-of-scope | — | Canvas SkillsService.getSkills uses the bundled @openhands/extensions catalog and the local /api/skills list with load_public:false. It does not call the SDK skills marketplace loader/endpoint whose default.json assumption is reported. |+| [OpenHands/software-agent-sdk#4664](https://github.com/OpenHands/software-agent-sdk/issues/4664) | out-of-scope | — | OpenHandsCloudWorkspace is a Python workspace object; Canvas uses TypeScript clients and Cloud runtime API transport, not that object or its cached Python connection fields. |+| [OpenHands/software-agent-sdk#4663](https://github.com/OpenHands/software-agent-sdk/issues/4663) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | Concurrent real glob/tool calls in distinct workspaces with the Python fallback active. |+| [OpenHands/software-agent-sdk#4661](https://github.com/OpenHands/software-agent-sdk/issues/4661) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | Two native task-enabled conversations with same-name project subagent definitions and a working model for delegation. |+| [OpenHands/software-agent-sdk#4660](https://github.com/OpenHands/software-agent-sdk/issues/4660) | out-of-scope | — | Canvas Model Router uses meta-profiles and buildRouterAtStartSystemSuffix to guide profile selection; it does not instantiate the Python RouterLLM or call its completion/acompletion methods. The reported SDK class path is not its implementation. |+| [OpenHands/software-agent-sdk#4659](https://github.com/OpenHands/software-agent-sdk/issues/4659) | out-of-scope | — | Canvas BashService.listOutputs reads every result page sorted by timestamp, and no Canvas consumer polls BashOutput with order__gt. The reported duplicate order values affect reconstruction clients using that cursor; the current Canvas logs consumer does not use the affected order contract. |+| [OpenHands/software-agent-sdk#4658](https://github.com/OpenHands/software-agent-sdk/issues/4658) | out-of-scope | — | The report is transactionality/resource unwind in POST /api/init for a deferred warm-pool coordinator. Canvas neither calls that init endpoint nor owns its staged runtime publication, and no Canvas-observable failure is given in the report; this is a coordinator/Agent Server lifecycle contract before the UI attaches. |+| [OpenHands/software-agent-sdk#4629](https://github.com/OpenHands/software-agent-sdk/issues/4629) | mapped | [F13.acp-credentials](../feature-map/F13-agent-profiles.md) | An empty run-owned Gemini OAuth file produced the already-signed-in banner on two fresh ACP forms. Provider auth priority and version migration were not exercised. |+| [OpenHands/software-agent-sdk#4613](https://github.com/OpenHands/software-agent-sdk/issues/4613) | mapped | [F10.custom-model](../feature-map/F10-llm-profiles.md) | A namespaced model such as openai/openai/example-model can have its inner provider prefix stripped again in Responses API and reach the provider as example-model. Reproduction needs an approved OpenAI-compatible endpoint serving the namespaced model and its exact stored API-mode settings; a loopback connection refusal cannot prove the outbound identifier. |+| [OpenHands/software-agent-sdk#4583](https://github.com/OpenHands/software-agent-sdk/issues/4583) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model invoking file-editor insert at EOF on a file with no trailing newline. |+| [OpenHands/software-agent-sdk#4575](https://github.com/OpenHands/software-agent-sdk/issues/4575) | mapped | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | A model key and controlled bounded workload/resource envelope for a real Canvas terminal run are unavailable. |+| [OpenHands/software-agent-sdk#4562](https://github.com/OpenHands/software-agent-sdk/issues/4562) | mapped | [F08.files-load-error](../feature-map/F08-workspace-files-and-changes.md) | A plain-HTTP non-loopback same-origin browser deployment with real workspace-cookie observation. |+| [OpenHands/software-agent-sdk#4558](https://github.com/OpenHands/software-agent-sdk/issues/4558) | mapped | [F06.error-events](../feature-map/F06-agent-activity.md) | A working encrypted saved fallback profile and real primary-model failure that triggers fallback resolution. |+| [OpenHands/software-agent-sdk#4540](https://github.com/OpenHands/software-agent-sdk/issues/4540) | mapped | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | Synthetic provider credentials and a GLM/Kimi turn emitting the raw tool_call wrapper. |+| [OpenHands/software-agent-sdk#4537](https://github.com/OpenHands/software-agent-sdk/issues/4537) | mapped | [F04.card](../feature-map/F04-conversation-list.md) | A working model with native TaskTool delegation holding the parent state lock during list reload. |+| [OpenHands/software-agent-sdk#4533](https://github.com/OpenHands/software-agent-sdk/issues/4533) | not-reproduced | [F13.active-drives-conversation](https://github.com/OpenHands/software-agent-sdk/issues/4533#issuecomment-6121790768) | Two seeded-default UI launches preserved the dummy key in runtime state and reached loopback connection refusal. This verifies key propagation, not successful provider authentication. |+| [OpenHands/software-agent-sdk#4532](https://github.com/OpenHands/software-agent-sdk/issues/4532) | mapped | [F06.confirmation-mode](../feature-map/F06-agent-activity.md) | A real confirmation-paused action persisted across backend restart, then approval and a working provider call. |+| [OpenHands/software-agent-sdk#4511](https://github.com/OpenHands/software-agent-sdk/issues/4511) | mapped | [F10.schema-validation](../feature-map/F10-llm-profiles.md) | Disabling prompt caching can still emit cache_control on preexisting cache-marked tool messages. The report only supplies a Python serializer reproduction; a Canvas check needs the exact persisted/tool-message precondition and a provider request that exposes the cache markers, with approved model access. |+| [OpenHands/software-agent-sdk#4498](https://github.com/OpenHands/software-agent-sdk/issues/4498) | mapped | [F05.context-window-meter](../feature-map/F05-composer.md) | A real condensation splitting an action/result pair, followed by the next provider request. |+| [OpenHands/software-agent-sdk#4432](https://github.com/OpenHands/software-agent-sdk/issues/4432) | out-of-scope | — | Fetch MCP runtime dependency pin request with no concrete Canvas reproduction or user-visible failure in the report. |+| [OpenHands/software-agent-sdk#4382](https://github.com/OpenHands/software-agent-sdk/issues/4382) | mapped | [F05.context-window-meter](../feature-map/F05-composer.md) | An authenticated ACP provider emitting cache/thought token usage or reset/stalled cumulative cost updates. |+| [OpenHands/software-agent-sdk#4363](https://github.com/OpenHands/software-agent-sdk/issues/4363) | mapped | [F19.refresh](../feature-map/F19-plugins.md) | Completed Update changed persisted requested_ref from v1 to null. Visible version stayed 1.0.0 in this cache, so a default-HEAD jump was not observed. |+| [OpenHands/software-agent-sdk#4093](https://github.com/OpenHands/software-agent-sdk/issues/4093) | mapped | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | With ACP Python 0.11, Gemini can lose the current/available model state from NewSessionResponse. Reproduction needs the affected ACP dependency and Gemini adapter plus an approved Gemini account; the profile's saved model field alone is not session model state. |+| [OpenHands/software-agent-sdk#4080](https://github.com/OpenHands/software-agent-sdk/issues/4080) | mapped | [F04.card](../feature-map/F04-conversation-list.md) | A copied persisted conversation with its original unregistered observation kind and missing tool-module registration. |+| [OpenHands/software-agent-sdk#4077](https://github.com/OpenHands/software-agent-sdk/issues/4077) | mapped | [F06.reload-mid-run](../feature-map/F06-agent-activity.md) | A working stream interrupted/retried or delivered to a controlled stalled subscriber, with delta/state-order capture. |+| [OpenHands/software-agent-sdk#4265](https://github.com/OpenHands/software-agent-sdk/issues/4265) | out-of-scope | — | Zentra fork patch maintenance, outside this Canvas repository and supported runtime. |+| [OpenHands/software-agent-sdk#4264](https://github.com/OpenHands/software-agent-sdk/issues/4264) | out-of-scope | — | Zentra fork vendor build/lib packaging, outside this Canvas repository and runtime. |+| [OpenHands/software-agent-sdk#4263](https://github.com/OpenHands/software-agent-sdk/issues/4263) | out-of-scope | — | Zentra fork air-gap/egress policy, outside this Canvas repository and supported runtime. |+| [OpenHands/software-agent-sdk#4260](https://github.com/OpenHands/software-agent-sdk/issues/4260) | mapped | [F23.run-linked-conversation / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Owned Cloud reviewer/webhook deployment, authorized disposable review target and model credentials are unavailable. |+| [OpenHands/software-agent-sdk#4252](https://github.com/OpenHands/software-agent-sdk/issues/4252) | not-reproduced | [F18.personal-skill-dirs](https://github.com/OpenHands/software-agent-sdk/issues/4252#issuecomment-6121791082) | A persistence-folder skill appeared in the UI and in available/activated skills in two actual initialized chats. This verifies discovery before the model reply, not historical Docker mounts. |+| [OpenHands/software-agent-sdk#4250](https://github.com/OpenHands/software-agent-sdk/issues/4250) | mapped | [F10.custom-model / F10.default-used](../feature-map/F10-llm-profiles.md) | A Cloud Workers AI configuration can fail on context-window validation or be routed as OpenAI with an invalid Cloudflare key. Reproduction needs a disposable OpenHands Cloud account, its Workers AI endpoint/key and the exact saved model configuration; the report supplies no complete configuration steps. |+| [OpenHands/software-agent-sdk#4248](https://github.com/OpenHands/software-agent-sdk/issues/4248) | out-of-scope | — | This unspecified legacy uv CLI report names execute_bash/security_risk but gives no Canvas/Agent Server version, UI entry point or reproduction. The current Canvas launcher is not that SDK CLI; no affected supported Canvas runtime consumer is identified. |+| [OpenHands/software-agent-sdk#4731](https://github.com/OpenHands/software-agent-sdk/issues/4731) | out-of-scope | — | TypeScript package bundler compatibility for require(ws); current Canvas browser event transport is separate and the frozen Canvas production build succeeds. |+| [OpenHands/software-agent-sdk#4247](https://github.com/OpenHands/software-agent-sdk/issues/4247) | mapped | [F10.custom-model](../feature-map/F10-llm-profiles.md) | An LM Studio configuration can fail with LLM Provider NOT provided although its models endpoint works. Reproduction needs the reachable LM Studio service and the exact model ID/provider-qualified config, which the report does not supply as text; the reported CLI/Docker SDK 1.16.1 installation is not this source stack. |+| [OpenHands/software-agent-sdk#4246](https://github.com/OpenHands/software-agent-sdk/issues/4246) | not-reproduced | [F17.agent-uses-server](https://github.com/OpenHands/software-agent-sdk/issues/4246#issuecomment-6121791434) | Two actual stdio timeouts left initialization after about 30 seconds, persisted the prompt/user message and reached loopback model error. This settles the current Local initialization loop only, not successful replies or historical Cloud/SSE behavior; sibling tools were lost. |+| [OpenHands/software-agent-sdk#4245](https://github.com/OpenHands/software-agent-sdk/issues/4245) | out-of-scope | — | Legacy docker.openhands.dev/openhands/openhands v0 webhook deployment, absent from current Agent Canvas stack. |+| [OpenHands/automation#565](https://github.com/OpenHands/automation/issues/565) | mapped | [F25.cloud-org-rows / F25.cloud-org-suspended](../feature-map/F25-backends-and-cloud.md) | Authenticated Cloud account/org-bound key and unauthorized-org fixture are unavailable. |+| [OpenHands/automation#557](https://github.com/OpenHands/automation/issues/557) | mapped | [F23.configuration / F23.run-linked-conversation](../feature-map/F23-automation-detail.md) | Saved bare qa-backfill-bare agent profile (tools=[], unique persona, closed-loopback qa-offline LLM), created prompt preset then PATCHed its selected agent_profile_id. Real UI Run now created a linked conversation whose launched_agent_profile is null, selected persona is absent and default tools are present; the selected profile was ignored before the deliberate closed-loopback LLM connection failure. |+| [OpenHands/automation#514](https://github.com/OpenHands/automation/issues/514) | mapped | [F03.cancel-in-flight-run / F23.activity-log / F23.run-logs-modal](../feature-map/F03-home.md) | Clicked Home owned worker Cancel run; toast Automation run cancelled and Activity Log CANCELLED. The python3 worker.py remains in this run launcher ancestry after cancellation, despite no provider or external worker calls. |+| [OpenHands/automation#472](https://github.com/OpenHands/automation/issues/472) | out-of-scope | — | Automation subject ownership and cancellation retry contract; Canvas does not submit subject-turn idempotency keys. |+| [OpenHands/automation#470](https://github.com/OpenHands/automation/issues/470) | out-of-scope | — | Automation subject-turn retry/idempotency API; Canvas does not submit subject turns or idempotency keys in its UI. |+| [OpenHands/automation#418](https://github.com/OpenHands/automation/issues/418) | out-of-scope | — | PR-review-bot merge gating, review attribution and operational audit; no concrete mapped Canvas UI behavior. |+| [OpenHands/automation#347](https://github.com/OpenHands/automation/issues/347) | mapped | [F21.card-grid / F21.overview-tiles](../feature-map/F21-automations-dashboard.md) | The owned 25-automation local dashboard completed run summaries with HTTP 200 and no 30 s stall on automation 1.19.3. This narrow live probe does not account for the issue’s missing batch-summary API/pool configuration contracts or withheld original lock workload. A failing dashboard recipe needs the original DB load/lock setup and redacted service-log/HAR timing evidence, linked to Canvas #16559. |+| [OpenHands/automation#343](https://github.com/OpenHands/automation/issues/343) | mapped | [F22.setup-tarball-create / F23.script-section / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Uploaded owned tarball scripts/setup.sh writes .qa-setup-marker, main.py requires marker; persisted setup_script_path scripts/setup.sh. UI Run now genuine BashOutput exits1 with QA_NESTED_SETUP_WAS_SKIPPED, proving nested setup ignored. |+| [OpenHands/automation#286](https://github.com/OpenHands/automation/issues/286) | mapped | [F23.run-status-polling](../feature-map/F23-automation-detail.md) | Owned multi-worker/replica automation deployment and authenticated Cloud sandbox verification context are unavailable. |+| [OpenHands/automation#246](https://github.com/OpenHands/automation/issues/246) | mapped | [F21.backend-unavailable / F21.list-error](../feature-map/F21-automations-dashboard.md) | The exact 404 URL, user action, version and local launcher/service configuration are missing from the report. |+| [OpenHands/automation#219](https://github.com/OpenHands/automation/issues/219) | mapped | [F23.run-status-polling](../feature-map/F23-automation-detail.md) | Owned disposable Cloud account at its sandbox cap with an interactive fixture is unavailable. |+| [OpenHands/automation#200](https://github.com/OpenHands/automation/issues/200) | out-of-scope | — | Automation KV REST request strictness and response contracts; no Canvas KV editor or consumer of these writes. |+| [OpenHands/automation#113](https://github.com/OpenHands/automation/issues/113) | mapped | [F23.run-linked-conversation](../feature-map/F23-automation-detail.md) | Affected automation/run/conversation ids, deployment version and genuine plugin-run credentials are unavailable. |++## Teardown and retained evidence++The JSON ledger records each run’s actual shutdown time and retained evidence count. Ledgers remain under `<verification-run-root>/<run-id>/evidence/ledger.jsonl`; the sanitized check ledger above includes their observations without live session URLs or private state. Selected reviewed media is included in this directory. Run owners delete their disposable fixtures or stop their private stack, restore altered keys/network/clock state, and stop bounded worker descendants before hand-off.++No authentication data, private logs or raw traces are published. No issue was closed and no product code changed. The accepted maintenance baseline remains unchanged.diff --git a/.agents/skills/verify-openhands/references/backfills/2026-10-16-main-refresh.json b/.agents/skills/verify-openhands/references/backfills/2026-10-16-main-refresh.jsonnew file mode 100644index 0000000000..bd7966ba65--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/2026-10-16-main-refresh.json@@ -0,0 +1,20311 @@+{+ "schema_version": 2,+ "cleanup": {+ "backfill-18334-refresh": {+ "launcher_stopped": true,+ "all_ports_closed": true,+ "private_purged": true,+ "retained_evidence_files": 70+ },+ "backfill-18334-settings-root": {+ "launcher_stopped": true,+ "all_ports_closed": true,+ "private_purged": true,+ "retained_evidence_files": 47+ }+ },+ "harness_limits": [+ "Known-bug JSON assertion refuses a nonzero api command before evaluating its fields; error-status recipes observe the real UI network record instead.",+ "ERR-trap failedCommand metadata may name its formatter; use each recipe, marked assertion and transcript for the actual action."+ ],+ "issue": "https://github.com/OpenHands/OpenHands/issues/18309",+ "target": "d9ecec24181a0dd7b83b5d87561916abb2fe0e8c",+ "live_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "inherited_maintenance_baseline": "1c75a05d587bcb502cd3f63d7e13eaa98ca84d53",+ "baseline_advanced_by_backfill": false,+ "historical_snapshot": "2026-10-12-open-bugs.json",+ "inventory_snapshots": {+ "OpenHands/OpenHands": {+ "captured_at": "2026-10-15T23:27:24.888074+00:00",+ "count": 211+ },+ "OpenHands/software-agent-sdk": {+ "captured_at": "2026-10-15T23:27:28.302757+00:00",+ "count": 199+ },+ "OpenHands/automation": {+ "captured_at": "2026-10-15T23:27:30.964305+00:00",+ "count": 14+ }+ },+ "counts": {+ "OpenHands/OpenHands": {+ "open_bugs": 211,+ "mapped": 177,+ "not-reproduced": 11,+ "out-of-scope": 23,+ "pending": 0+ },+ "OpenHands/software-agent-sdk": {+ "open_bugs": 199,+ "mapped": 114,+ "not-reproduced": 7,+ "out-of-scope": 78,+ "pending": 0+ },+ "OpenHands/automation": {+ "open_bugs": 14,+ "mapped": 10,+ "not-reproduced": 0,+ "out-of-scope": 4,+ "pending": 0+ }+ },+ "new_inventory_issues": 96,+ "historical_issues_now_closed": 34,+ "additional_closed_map_references": 6,+ "accounting_complete": true,+ "problems": [],+ "runtimes": [+ {+ "runId": "backfill-18334-refresh",+ "revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "buildId": "c0efc42e518128f2",+ "startedAt": "2026-10-15T23:34:41.836Z",+ "readyAt": "2026-10-15T23:34:52.435Z",+ "ports": {+ "ingress": 18920,+ "agentServer": 18921,+ "automation": 18922,+ "frontend": 18923,+ "vscode": 19921+ },+ "pins": {+ "agentServer": "1.55.0",+ "agentCanvas": "1.27.1",+ "automation": "1.20.0"+ },+ "versions": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0"+ }+ },+ {+ "runId": "backfill-18334-settings-root",+ "revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "buildId": "c0efc42e518128f2",+ "startedAt": "2026-10-16T04:15:49.835Z",+ "readyAt": "2026-10-16T04:16:12.464Z",+ "ports": {+ "ingress": 18940,+ "agentServer": 18941,+ "automation": 18942,+ "frontend": 18943,+ "vscode": 19941+ },+ "pins": {+ "agentServer": "1.55.0",+ "agentCanvas": "1.27.1",+ "automation": "1.20.0"+ },+ "versions": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0"+ }+ }+ ],+ "verification_scope": "Current source/pinned-consumer reconciliation plus the explicitly recorded live checks. Historical notes keep their original dates; no whole-map live audit or baseline/fix proof is claimed. No real model credentials or paid inference.",+ "issues": [+ {+ "repository": "OpenHands/OpenHands",+ "number": 18480,+ "title": "[Bug]: Repository picker shows only one Azure DevOps organization's repositories",+ "url": "https://github.com/OpenHands/OpenHands/issues/18480",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F03",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "feature_ids": [+ "F03.open-repository-dialog"+ ],+ "basis": "current-source-review",+ "reason": "Azure DevOps Cloud picker uses direct listing rather than organization installations, omitting other organizations.",+ "source_refs": [+ "src/utils/utils.ts",+ "src/hooks/query/use-git-repositories.ts",+ "src/hooks/query/use-app-installations.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Real Enterprise Azure DevOps sign-in belonging to two organizations with distinct repositories; original report used mocks only.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18461,+ "title": "[Bug]: Switching an automation schedule to Custom drops the chosen preset (always 0 9 * * *)",+ "url": "https://github.com/OpenHands/OpenHands/issues/18461",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F28",+ "feature_ids": [+ "F28.schedule-presets"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F28-automation-setup-page.md",+ "basis": "live-reproduced",+ "reason": "Fresh create-mode replay still drops Weekly Monday 10:30 when switching to Custom: textbox reads 0 9 * * * instead of 30 10 * * 1. The existing Oct15 failure remains current on this target.",+ "current_result": "fail",+ "fresh_reproduction_claim": true,+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "Automate \u2192 Add Automation \u2192 Create Automation. Choose Weekly (Monday is selected), set At 10:30, then choose Custom. Read automation-setup-custom-schedule. Do not create or dispatch the automation."+ ],+ "artifacts": [+ "evidence/F28.schedule-presets/current-weekly-custom.png",+ "evidence/F28.schedule-presets/current-weekly-custom.json"+ ],+ "verification_status": "reproduced",+ "verification_detail": "verified-current-failure"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18445,+ "title": "[Bug]: Completed conversation opens with empty history on cloud \u2014 silent empty-page fallback on events/search",+ "url": "https://github.com/OpenHands/OpenHands/issues/18445",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.load-older-history",+ "F06.reconnect"+ ],+ "basis": "current-source-review",+ "reason": "Cloud initial history failure is swallowed as empty page; successful runtime count can then advance after_seq past unseeded history.",+ "source_refs": [+ "src/api/event-service/event-service.api.ts",+ "src/hooks/query/use-conversation-history.ts",+ "src/contexts/conversation-websocket-context.tsx"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Cloud account and owned conversation whose genuine App API events/search fails while runtime events/count succeeds; Local throws and is a different path.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18434,+ "title": "[Bug]: DigitalOcean sign-in always fails with \"session key required\" when the session key was typed on the API key screen (LAN bind, Docker)",+ "url": "https://github.com/OpenHands/OpenHands/issues/18434",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F25",+ "feature_ids": [+ "F25.digitalocean-sign-in"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "existing-main-note",+ "reason": "Upstream F25 already records the Oct14 typed-session-key bridge failure. Current client still reads only baked headers. Requires a separate owned non-baked-key public origin and real API-key-screen entry; the default baked-key stack does not exercise it.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Upstream F25 already records the Oct14 typed-session-key bridge failure. Current client still reads only baked headers. Requires a separate owned non-baked-key public origin and real API-key-screen entry; the default baked-key stack does not exercise it.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18432,+ "title": "[Bug]: The agent's secret request dialog is wider than a phone screen, cutting off its title, fields and Cancel",+ "url": "https://github.com/OpenHands/OpenHands/issues/18432",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F14",+ "feature_ids": [+ "F14.agent-request"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F14-secrets.md",+ "basis": "existing-main-note",+ "reason": "Current main already records the phone request-secret dialog overflow with an executable known-bug block and a dated2026-10-14 observation. Preserve that date; no Oct16 reproduction completed.",+ "source_refs": [+ "src/components/features/conversation/secret-request-modal.tsx",+ ".agents/skills/verify-openhands/references/feature-map/F14-secrets.md"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "A genuine request_secret client-tool invocation on a new OpenHands conversation; needs an approved model turn unless another real client-tool trigger is provided.",+ "map_note_status": "already-linked-main",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18421,+ "title": "[Bug]: Chat attachment upload fails with 422 \"path must be inside the conversation workspace\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/18421",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F05",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "feature_ids": [+ "F05.attach-files"+ ],+ "basis": "live-reproduced",+ "reason": "Actual Home New Worktree send posted attachment to original repository path and returned422 path must be inside conversation workspace; created runtime workspace is /tmp/conversation-worktrees/<id>/qa-surface-attachment. Same file sent inside that existing conversation uploaded to worktree path200. Only dummy refused-loopback profile used.",+ "source_refs": [+ "src/hooks/chat/use-chat-attachment-upload.ts",+ "src/api/conversation-file-upload.api.ts",+ "src/api/workspace-upload-path.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": {+ "repository": "OpenHands/OpenHands",+ "number": 18421,+ "result": "fail",+ "actual": "Actual Home New Worktree send posted attachment to original repository path and returned422 path must be inside conversation workspace; created runtime workspace is /tmp/conversation-worktrees/<id>/qa-surface-attachment. Same file sent inside that existing conversation uploaded to worktree path200. Only dummy refused-loopback profile used.",+ "expected": "Home attachment resolves against newly created conversation worktree and uploads200 before message send.",+ "evidence": [+ "evidence/F05.attach-files/home-new-worktree-refresh-18421.png",+ "evidence/F05.attach-files/existing-worktree-control-18421-2.png"+ ],+ "steps": [+ "fixture git-repo --name qa-surface-attachment; fixture file --name qa-surface-note.txt.",+ "Home \u2192 workspace open qa-surface-attachment \u2192 workspace-mode-selector \u2192 new_worktree.",+ "Add file using actual chooser, fill prompt, click Home Send. Network scoped file/upload original-path422; toast preserves detail.",+ "On created conversation add same file and send bounded offline prompt. Upload uses actual worktree path200."+ ],+ "at": "2026-10-16T04:52:27.193836+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "feature_ids": [+ "F05.attach-files"+ ],+ "known_bug_contract": {+ "id": "F05.attach-files",+ "name": "home-new-worktree",+ "status": "xfail",+ "sha": "081b1850a6764519",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "422",+ "bugShows": "422",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"422\"",+ "actual": "422",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F05.attach-files-home-new-worktree-1792126644776-1.json"+ }+ },+ "live_recipe": "Execute existing F05.attach-files home-new-worktree block using owned repository/file and closed-loopback profile; upload occurs before provider inference.",+ "known_bug_contract": {+ "id": "F05.attach-files",+ "name": "home-new-worktree",+ "status": "xfail",+ "sha": "081b1850a6764519",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "422",+ "bugShows": "422",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"422\"",+ "actual": "422",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F05.attach-files-home-new-worktree-1792126644776-1.json"+ },+ "current_result": "fail",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-refresh",+ "steps": [+ "fixture git-repo --name qa-surface-attachment; fixture file --name qa-surface-note.txt.",+ "Home \u2192 workspace open qa-surface-attachment \u2192 workspace-mode-selector \u2192 new_worktree.",+ "Add file using actual chooser, fill prompt, click Home Send. Network scoped file/upload original-path422; toast preserves detail.",+ "On created conversation add same file and send bounded offline prompt. Upload uses actual worktree path200."+ ],+ "artifacts": [+ "evidence/F05.attach-files/home-new-worktree-refresh-18421.png",+ "evidence/F05.attach-files/existing-worktree-control-18421-2.png"+ ],+ "observation": "Actual Home New Worktree send posted attachment to original repository path and returned422 path must be inside conversation workspace; created runtime workspace is /tmp/conversation-worktrees/<id>/qa-surface-attachment. Same file sent inside that existing conversation uploaded to worktree path200. Only dummy refused-loopback profile used.",+ "limit": null,+ "no_successful_model_inference": true+ },+ "run_id": "backfill-18334-refresh",+ "steps": [+ "fixture git-repo --name qa-surface-attachment; fixture file --name qa-surface-note.txt.",+ "Home \u2192 workspace open qa-surface-attachment \u2192 workspace-mode-selector \u2192 new_worktree.",+ "Add file using actual chooser, fill prompt, click Home Send. Network scoped file/upload original-path422; toast preserves detail.",+ "On created conversation add same file and send bounded offline prompt. Upload uses actual worktree path200."+ ],+ "artifacts": [+ "evidence/F05.attach-files/home-new-worktree-refresh-18421.png",+ "evidence/F05.attach-files/existing-worktree-control-18421-2.png"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18418,+ "title": "[Bug]: Creating a conversation always fails: the sandbox agent cannot find the plugin schema.",+ "url": "https://github.com/OpenHands/OpenHands/issues/18418",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F03",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "feature_ids": [+ "F03.create-error-toast",+ "F03.launch-with-plugin"+ ],+ "basis": "current-source-review",+ "reason": "Hosted1.50.1 runtime fails during skills/plugins setup when bundled plugin-schema is absent.",+ "source_refs": [+ "src/api/cloud/conversation-service.api.ts",+ "src/hooks/query/use-task-polling.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Affected hosted runtime1.50.1/PyInstaller image and original plugin configuration plus Cloud account; Local1.55 is not equivalent.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18390,+ "title": "[Bug]: Activating a newly added backend from a conversation page sometimes asks it for the old backend's conversation id",+ "url": "https://github.com/OpenHands/OpenHands/issues/18390",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F25",+ "feature_ids": [+ "F25.add-agent-server",+ "F25.switch-backend"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "existing-main-note",+ "reason": "Current-main F25 already preserves historical first-activation race observations; added the current issue link. Recheck at least15 clean/failed-then-clean new-backend activations from a real conversation, observing old-id traffic. A single clean switch cannot clear this intermittent report.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Current-main F25 already preserves historical first-activation race observations; added the current issue link. Recheck at least15 clean/failed-then-clean new-backend activations from a real conversation, observing old-id traffic. A single clean switch cannot clear this intermittent report.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18362,+ "title": "[Bug]: Fail Canvas startup when the automation backend exits before readiness",+ "url": "https://github.com/OpenHands/OpenHands/issues/18362",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F26",+ "feature_ids": [+ "F26.automation-startup-failure"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "existing-main-note",+ "reason": "Current-main F26 records Oct13 startup-exit failure and its owned failing-version launcher recipe. Preserve those dates. A new bounded secondary launcher with private HOME/state and free ports is required; it must not disrupt the shared main stack.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Current-main F26 records Oct13 startup-exit failure and its owned failing-version launcher recipe. Preserve those dates. A new bounded secondary launcher with private HOME/state and free ports is required; it must not disrupt the shared main stack.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18341,+ "title": "Automation Edit shows starter Python instead of the saved custom automation bundle",+ "url": "https://github.com/OpenHands/OpenHands/issues/18341",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F21",+ "feature_ids": [+ "F21.edit",+ "F23.edit-open",+ "F28.kind-toggle"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "basis": "existing-main-note",+ "reason": "Current-main maps the saved-bundle normalization defect in F21 and crosslinks F23/F28. The current form session still drops existingCustomTarballPath. Recheck a reviewed uploaded bundle through dashboard Edit and an unrelated form update, rather than relying on a direct cold URL.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Current-main maps the saved-bundle normalization defect in F21 and crosslinks F23/F28. The current form session still drops existingCustomTarballPath. Recheck a reviewed uploaded bundle through dashboard Edit and an unrelated form update, rather than relying on a direct cold URL.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18324,+ "title": "Use readable theme focus and valid legacy scrim opacity",+ "url": "https://github.com/OpenHands/OpenHands/issues/18324",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F25",+ "feature_ids": [+ "F25.add-backend-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "existing-main-note",+ "reason": "At the original b34ac59 snapshot, backend tabs/guidance use outline-blue-300. F25 retains Oct13 focus evidence. F09 secondary consumer is the exported legacy SettingsNavigation/SettingsMobileDrawer build path; the current route uses SettingsMobileHub, so that consumer needs compiled-CSS/export evidence rather than a claimed live phone route.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "secondary_families": [+ "F09"+ ],+ "secondary_consumer": "Public exported legacy SettingsNavigation \u2192 SettingsMobileDrawer; current route uses SettingsMobileHub. Compiled-CSS/build contract, not current phone navigation.",+ "verification_status": "resolved-after-snapshot",+ "prerequisite": "At the original b34ac59 snapshot, backend tabs/guidance use outline-blue-300. F25 retains Oct13 focus evidence. F09 secondary consumer is the exported legacy SettingsNavigation/SettingsMobileDrawer build path; the current route uses SettingsMobileHub, so that consumer needs compiled-CSS/export evidence rather than a claimed live phone route.",+ "post_snapshot_update": "OpenHands/OpenHands#18324 closure is recorded separately; original open snapshot state preserved.",+ "post_snapshot_issue_state": "closed",+ "post_snapshot_disposition": "Resolved in later main; obsolete issue note removed from the current map.",+ "post_snapshot_closed_at": "2026-10-16T04:12:07Z"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18302,+ "title": "[Bug]: Onboarding's LLM step overwrites an existing LLM profile that has the model-derived name",+ "url": "https://github.com/OpenHands/OpenHands/issues/18302",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F01",+ "feature_ids": [+ "F01.onboarding-repeat-endpoint"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md",+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18301,+ "title": "[Bug]: Preserve Automation KV encryption across session-key rotation",+ "url": "https://github.com/OpenHands/OpenHands/issues/18301",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.session-key-rotated",+ "F26.session-key-persist"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Wrote and read an owned Automation KV marker (HTTP 201/200), then control-openhands restart --rotate-key with no AUTOMATION_KV_SECRET override. Fresh-auth KV GET now returns HTTP 500 instead of the prior marker.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18299,+ "title": "[Bug]: migration 029 errors with `duplicate column name: consecutive_auth_failures` on SQLite",+ "url": "https://github.com/OpenHands/OpenHands/issues/18299",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Stopped only the owned automation service, backed up its DB, reset alembic_version from 032 to 028 while consecutive_auth_failures exists, restarted the real launcher. Migration duplicate-column signature and automation unavailability occur while launcher stays alive. Private original DB restored afterward.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18287,+ "title": "[Bug]: Repeated cards for the same skill source when local and bundled skill names collide",+ "url": "https://github.com/OpenHands/OpenHands/issues/18287",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F18",+ "feature_ids": [+ "F18.card",+ "F18.facets"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-settings.tsx",+ "src/components/features/skills/",+ "src/constants/skills-docs.ts",+ "src/hooks/use-skill-enablement.ts",+ "src/utils/skill-enablement.ts",+ "src/hooks/use-launch-skill-in-chat.ts",+ "src/api/skills-service.ts",+ "src/components/features/chat/skill-install-restart-banner.tsx",+ "src/utils/skill-install-events.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/skills/skill-type-badge.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "line": 84+ },+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "line": 90+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18234,+ "title": "[Bug]: Pinned automation card on Home hides the run's task summary when the run has a conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/18234",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.home-run-outcome-health"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18233,+ "title": "[Bug]: Escape on the closed backend selector clears it: the backend name becomes a placeholder and the status dot disappears",+ "url": "https://github.com/OpenHands/OpenHands/issues/18233",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.selector-dropdown"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Historical reproduced note dated 2026-10-09 already describes closed-menu Escape clearing the active backend display; source Dropdown and backend selector still own the path.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18210,+ "title": "[Bug]: Home shows dead automation link and manifest errors when automation interface is absent",+ "url": "https://github.com/OpenHands/OpenHands/issues/18210",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F21",+ "feature_ids": [+ "F21.manifest-404"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "basis": "retained-historical-note",+ "reason": "Existing missing-manifest recipe and note describe the Home link/toasts; preserved source revision and prior evidence date.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18204,+ "title": "[Bug]: Provider connection menu does not focus its first action",+ "url": "https://github.com/OpenHands/OpenHands/issues/18204",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F11",+ "feature_ids": [+ "F11.row-menu-focus"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "basis": "live-reproduced",+ "reason": "Reproduced twice in existing Chrome on current pins: opening provider connection menu kept focus on provider-connection-menu-trigger; ArrowDown did not move it.",+ "source_refs": [+ "src/components/features/settings/llm-profiles/provider-connection-actions-menu.tsx",+ ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "pins": {+ "agentServer": "1.55.0",+ "automation": "1.20.0",+ "agentCanvas": "1.27.1"+ },+ "result": "fail",+ "attempts": 2,+ "observations": [+ {+ "active": "provider-connection-menu-trigger",+ "menu": [+ "Bulk add",+ "Edit",+ "Delete"+ ],+ "afterArrowDown": "provider-connection-menu-trigger"+ },+ {+ "active": "provider-connection-menu-trigger"+ }+ ],+ "steps": [+ "await tab.goto('http://127.0.0.1:18920/settings/llm')",+ "await tab.playwright.getByRole('button',{name:'Provider connection menu',exact:true}).click()",+ "await tab.playwright.evaluate(()=>document.activeElement?.getAttribute('data-testid'))",+ "await tab.playwright.getByRole('menu').innerText()",+ "await tab.playwright.locator(':focus').press('ArrowDown')",+ "await tab.playwright.evaluate(()=>document.activeElement?.getAttribute('data-testid'))",+ "await tab.playwright.locator(':focus').press('Escape')",+ "Repeat menu click/focused-element readback once, then Escape"+ ],+ "arrange": {+ "provider_connection": {+ "display_name": "QA_REFRESH_CONN",+ "provider": "openai",+ "api_key": "<disposable dummy only>",+ "base_url": "http://127.0.0.1:1/v1"+ }+ },+ "limitations": "No provider authentication/inference; this is keyboard-focus behavior.",+ "artifacts": [+ "evidence/F11.row-menu-focus/oct16-current-menu-focus.json"+ ],+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-refresh"+ },+ "map_note_status": "current Known failure; existing Chrome observations and ledger recorded",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "line": 63+ }+ ],+ "map_note_present": true,+ "current_result": "fail",+ "run_id": "backfill-18334-refresh",+ "steps": [+ "await tab.goto('http://127.0.0.1:18920/settings/llm')",+ "await tab.playwright.getByRole('button',{name:'Provider connection menu',exact:true}).click()",+ "await tab.playwright.evaluate(()=>document.activeElement?.getAttribute('data-testid'))",+ "await tab.playwright.getByRole('menu').innerText()",+ "await tab.playwright.locator(':focus').press('ArrowDown')",+ "await tab.playwright.evaluate(()=>document.activeElement?.getAttribute('data-testid'))",+ "await tab.playwright.locator(':focus').press('Escape')",+ "Repeat menu click/focused-element readback once, then Escape"+ ],+ "artifacts": [+ "evidence/F11.row-menu-focus/oct16-current-menu-focus.json"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18196,+ "title": "[Bug]: Canvas App trust confirmation lets keyboard focus activate background Update",+ "url": "https://github.com/OpenHands/OpenHands/issues/18196",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F20",+ "feature_ids": [+ "F20.enable-confirm"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "basis": "live-check",+ "reason": "Fresh keyboard replay on the reviewed local demo app focuses Cancel when trust confirmation opens. Tab and Shift+Tab wrap only between Cancel and Enable trusted app. Escape and keyboard Cancel close it and restore the still-disabled Enable switch. No background action was observed. The old background-focus failure does not reproduce; pending-mutation containment was not exercised.",+ "current_result": "pass",+ "fresh_reproduction_claim": false,+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "Customize \u2192 Apps \u2192 Add app; install the reviewed repository src/fixtures/canvas-extensions/demo-page disabled.",+ "Focus Enable switch via locator keyboard Space. Read active element; Tab Cancel\u2192Enable, Tab Enable\u2192Cancel, Shift+Tab Cancel\u2192Enable. Escape closes and returns focus to Enable (aria-checked=false).",+ "Space again, Enter on initial Cancel; read restored Enable focus and disabled state. Uninstall only the owned demo through its confirmation. No enable or Update action sent by this drive."+ ],+ "artifacts": [+ "evidence/F20.enable-confirm/current-focus-trap.png",+ "evidence/F20.enable-confirm/current-focus-trap.json"+ ],+ "verification_status": "not-reproduced",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/18196#issuecomment-6208219932",+ "comment_id": 6208219932,+ "refresh_published_at": "2026-10-16T04:40:10Z",+ "body_sha256": "a64eab8f9d1f214ead0613ad8ae90736f4ed048ad58afebac271b18793fe79eb"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18161,+ "title": "[Bug]: Slack MCP credential probe reports any failed tool call (e.g. \"fetch failed\") as \"Credential check failed\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/18161",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F17",+ "feature_ids": [+ "F17.credential-probe"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18160,+ "title": "[Bug]: Recovery gate (stored backend answers 503) shows two generic \"An error occurred\" toasts from the free-models hydrator",+ "url": "https://github.com/OpenHands/OpenHands/issues/18160",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.frontend-only-returning"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Existing reproduced 2026-10-08 recovery-gate hydrator/toast note preserved.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18155,+ "title": "[Bug]: Missing-runtime conversations shown as Archived but never hidden, chipped, or unarchivable",+ "url": "https://github.com/OpenHands/OpenHands/issues/18155",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.show-archived-unarchive"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "known-issue",+ "reason": "A Cloud sandbox or local runtime that becomes MISSING and non-resumable; ordinary local pause remains resumable.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A Cloud sandbox or local runtime that becomes MISSING and non-resumable; ordinary local pause remains resumable.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18136,+ "title": "[Bug]: Connect Repo on a cloud conversation does not save the repository on the server",+ "url": "https://github.com/OpenHands/OpenHands/issues/18136",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.cloud-only"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "An authenticated Cloud sandbox/repository provider and a second fresh browser to compare server repository metadata.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Cloud sandbox/repository provider and a second fresh browser to compare server repository metadata.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18123,+ "title": "[Bug]: Mock-LLM cross-connect test: the sidebar cmd-click sometimes opens no tab, and the test hangs for 6 minutes (2 of 9 main runs)",+ "url": "https://github.com/OpenHands/OpenHands/issues/18123",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Intermittent mock-LLM CI tab-opening wait and diagnostics; report has no reproduction outside the test runner and requests bounded test failure/artifacts. The current user-visible cmd-click feature remains mapped in F04, without claiming the CI flake as a reproduced product bug.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18117,+ "title": "[Bug]: Inline thinking leaks into the agent chat bubble",+ "url": "https://github.com/OpenHands/OpenHands/issues/18117",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.thinking"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A live model that emits a leading inline think block on a tool-call turn, plus its finalized ActionEvent.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A live model that emits a leading inline think block on a tool-call turn, plus its finalized ActionEvent.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18108,+ "title": "[Bug]: Files tab shows \"No files in workspace\" on cloud when the agent writes outside /workspace/project",+ "url": "https://github.com/OpenHands/OpenHands/issues/18108",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-tree"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "An authenticated Enterprise Cloud sandbox, no repository selected, with files outside its default working directory.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Enterprise Cloud sandbox, no repository selected, with files outside its default working directory.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18063,+ "title": "[Bug]: Composer \"More input actions\" Model submenu runs off the right and bottom edges at narrow widths (third profile and LLM Profiles link can't be clicked or tapped)",+ "url": "https://github.com/OpenHands/OpenHands/issues/18063",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.overflow-menu"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "See the issue-specific prerequisite in the linked Known issue note.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18062,+ "title": "[Bug]: Automations: a deleted automation stays on the dashboard, home page and detail page after Run now fails with \"Automation not found\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/18062",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F21.run-now-error",+ "F23.run-now-error"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Existing dashboard and detail stale-automation notes and recipes preserved; no fresh reproduction claimed.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18029,+ "title": "[Bug]: Locked-to-cloud Agent Canvas shows 'No backend configured' after onboarding despite a healthy configured backend",+ "url": "https://github.com/OpenHands/OpenHands/issues/18029",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.locked-cloud"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Affected locked-to-cloud deployment, authenticated Cloud session and stale-document/alternate-origin evidence are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected locked-to-cloud deployment, authenticated Cloud session and stale-document/alternate-origin evidence are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17959,+ "title": "[Bug]: Automation templates: a search with no matches shows a blank page, and cards say \"1 MCPs to connect\" for optional integrations",+ "url": "https://github.com/OpenHands/OpenHands/issues/17959",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F22",+ "feature_ids": [+ "F22.templates-search",+ "F22.template-card"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md",+ "basis": "retained-historical-note",+ "reason": "Both no-result template search and optional-integration pill are already recorded; preserved existing notes.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "See the issue-specific prerequisite in the linked Known issue note."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17953,+ "title": "[Bug]: Backends and Cloud sign-in: Escape closes the whole Manage backends stack, raw 404 toast on shared pages, stray about:blank pop-ups",+ "url": "https://github.com/OpenHands/OpenHands/issues/17953",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.edit-escape",+ "F25.shared-conversation-view",+ "F25.cloud-device-flow"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "All three reported failures already have existing map notes and drive recipes.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17949,+ "title": "[Bug]: Launcher: --allow-lan-session-key is ignored, the port-in-use error prints twice with a stack trace, and backend-only runtime_services lists a frontend",+ "url": "https://github.com/OpenHands/OpenHands/issues/17949",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.host-bind-lan-optin",+ "F26.cli-port-in-use",+ "F26.backend-only"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "All three launcher defects already mapped; retained reproduced 2026-10-08 flag note.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17948,+ "title": "[Bug]: Automation detail and edit: schedule edit silently dropped, wrong run label, slow not-found page, tarball saved as .tar",+ "url": "https://github.com/OpenHands/OpenHands/issues/17948",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.edit-time-cleared",+ "F23.run-no-conversation",+ "F23.not-found",+ "F23.tarball"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "All four automation-detail defects already recorded in Gotchas and corresponding recipes.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17946,+ "title": "[Bug]: Launcher never creates TMUX_TMPDIR, so local Agent Canvas instances share one tmux server and reset each other's terminals",+ "url": "https://github.com/OpenHands/OpenHands/issues/17946",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Existing npm tmux-isolation known issue preserved; verification launcher intentionally supplies isolation and is not a product-fix claim.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "See the issue-specific prerequisite in the linked Known issue note."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17941,+ "title": "[Bug]: Skill and MCP card toggles can swallow the first click, and keep the new state after a failed save",+ "url": "https://github.com/OpenHands/OpenHands/issues/17941",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F18",+ "feature_ids": [+ "F18.toggle",+ "F18.toggle-error"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-settings.tsx",+ "src/components/features/skills/",+ "src/constants/skills-docs.ts",+ "src/hooks/use-skill-enablement.ts",+ "src/utils/skill-enablement.ts",+ "src/hooks/use-launch-skill-in-chat.ts",+ "src/api/skills-service.ts",+ "src/components/features/chat/skill-install-restart-banner.tsx",+ "src/utils/skill-install-events.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/skills/skill-type-badge.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "not-run"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17932,+ "title": "[Bug]: Settings save retries a 422 validation error three times before showing it",+ "url": "https://github.com/OpenHands/OpenHands/issues/17932",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F15",+ "feature_ids": [+ "F15.save-errors"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/condenser-settings.tsx",+ "src/routes/agent-context-settings.tsx",+ "src/routes/verification-settings.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/utils/sdk-settings-schema.ts"+ ],+ "changed_canvas_source_paths": [+ "src/utils/sdk-settings-schema.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "not-run"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17927,+ "title": "[Bug]: A failed Save on Settings > Application discards the user's unsaved edits",+ "url": "https://github.com/OpenHands/OpenHands/issues/17927",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F16",+ "feature_ids": [+ "F16.save-states"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F16-application-settings.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/app-settings.tsx",+ "src/components/features/settings/app-settings/",+ "src/components/features/settings/settings-switch.tsx",+ "src/themes/color-theme/",+ "src/utils/transcription-endpoint-storage.ts",+ "src/api/settings-service/settings-service.api.ts",+ "src/services/telemetry.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/settings-switch.tsx",+ "src/routes/app-settings.tsx",+ "src/services/telemetry.ts",+ "src/themes/color-theme/definitions/light-plus.ts",+ "src/themes/color-theme/definitions/solarized-light.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "not-run"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17922,+ "title": "[Bug]: MCP library lists entries that cannot be installed on a local backend; their install dialog is empty and Install does nothing",+ "url": "https://github.com/OpenHands/OpenHands/issues/17922",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F17",+ "feature_ids": [+ "F17.install-unsupported"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 129+ }+ ],+ "map_note_present": true,+ "verification_status": "not-run"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17919,+ "title": "[Bug]: \"Delete all\" conversations deletes only the loaded first page and states a smaller count",+ "url": "https://github.com/OpenHands/OpenHands/issues/17919",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.delete-all"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "See the issue-specific prerequisite in the linked Known issue note.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17899,+ "title": "[Bug]: Git user name and email from Settings > Application are never applied to agent commits",+ "url": "https://github.com/OpenHands/OpenHands/issues/17899",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F16",+ "feature_ids": [+ "F16.git-identity"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F16-application-settings.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/app-settings.tsx",+ "src/components/features/settings/app-settings/",+ "src/components/features/settings/settings-switch.tsx",+ "src/themes/color-theme/",+ "src/utils/transcription-endpoint-storage.ts",+ "src/api/settings-service/settings-service.api.ts",+ "src/services/telemetry.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/settings-switch.tsx",+ "src/routes/app-settings.tsx",+ "src/services/telemetry.ts",+ "src/themes/color-theme/definitions/light-plus.ts",+ "src/themes/color-theme/definitions/solarized-light.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "not-run"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17898,+ "title": "[Bug]: VITE_DO_NOT_TRACK=1 still initializes PostHog and loads remote scripts from the telemetry host",+ "url": "https://github.com/OpenHands/OpenHands/issues/17898",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F16",+ "feature_ids": [+ "F16.analytics"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F16-application-settings.md",+ "basis": "live-not-reproduced",+ "reason": "Two real opt-in Save/reload cycles on the hard-DNT build produced zero PostHog/telemetry-host requests and no PostHog storage keys. The second reloaded form showed consent checked; consent was then restored off. This covers the hard VITE_DO_NOT_TRACK build, not a browser-only DNT preference.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17898#issuecomment-6121789760",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-reproduced",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "passes": 2,+ "steps": [+ "control-openhands browser goto /settings/app",+ "control-openhands browser network --external --filter 'posthog|z.openhands.dev'",+ "control-openhands browser storage",+ "control-openhands browser click 'testid=app-settings-screen >> text=Send anonymous usage data'",+ "control-openhands browser click 'testid=submit-button'",+ "control-openhands browser reload",+ "control-openhands browser network --external --filter 'posthog|z.openhands.dev'",+ "control-openhands browser storage",+ "control-openhands browser click 'testid=app-settings-screen >> text=Send anonymous usage data'",+ "control-openhands browser click 'testid=submit-button'",+ "control-openhands doctor"+ ],+ "evidence": [],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Two network snapshots reported zero PostHog/telemetry-host requests around the real consent-switch Save/reload action; browser storage had no PostHog key. Consent was restored off. The check did not separately record the persisted consent-true field, and covers the hard VITE_DO_NOT_TRACK build only.",+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "build_flag": "VITE_DO_NOT_TRACK=1 at build time (the existing frozen harness build; fingerprint d465c70f971b4d8e); runtime DO_NOT_TRACK=1. A browser-only DNT flag is not equivalent."+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/app-settings.tsx",+ "src/components/features/settings/app-settings/",+ "src/components/features/settings/settings-switch.tsx",+ "src/themes/color-theme/",+ "src/utils/transcription-endpoint-storage.ts",+ "src/api/settings-service/settings-service.api.ts",+ "src/services/telemetry.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/settings-switch.tsx",+ "src/routes/app-settings.tsx",+ "src/services/telemetry.ts",+ "src/themes/color-theme/definitions/light-plus.ts",+ "src/themes/color-theme/definitions/solarized-light.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "HardDNT is a build-time condition; current main also documents a distinct explicit non-DNT consent-reversal flow. Do not use either historicalobservation to infer the other build."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Launch this target with the real hard VITE_DO_NOT_TRACK build; doctor before driving.",+ "control-openhands browser goto /settings/app --run <run>",+ "Click testid=app-settings-screen >> text=Send anonymous usage data; click testid=submit-button; wait for Settings saved; reload.",+ "Read the reloaded form and control-openhands browser network --external --filter \"posthog|z.openhands.dev\"; inspect browser storage for PostHog keys.",+ "Restore consent off and save; repeat the opt-in Save/reload check, confirm checked consent persisted; read network/storage again.",+ "Restore consent off and save; doctor. No successful model call was needed."+ ],+ "observations": {+ "network_requests_matching_telemetry": 0,+ "posthog_storage_keys": 0,+ "consent_checked_after_reload": true,+ "consent_restored_off": true+ },+ "artifacts": [+ "evidence/F16.analytics/oct16-hard-dnt-after-replay.png"+ ],+ "no_successful_model_inference": true+ },+ "current_result": "pass",+ "fresh_live_outcome": "not-reproduced",+ "comment_id": 6121789760,+ "refresh_published_at": "2026-10-16T04:40:11Z",+ "body_sha256": "e0dc7a1912a759802786aaef41fce79c5ef0b48494c1d0c6de4cbea40c8a0893",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "Launch this target with the real hard VITE_DO_NOT_TRACK build; doctor before driving.",+ "control-openhands browser goto /settings/app --run <run>",+ "Click testid=app-settings-screen >> text=Send anonymous usage data; click testid=submit-button; wait for Settings saved; reload.",+ "Read the reloaded form and control-openhands browser network --external --filter \"posthog|z.openhands.dev\"; inspect browser storage for PostHog keys.",+ "Restore consent off and save; repeat the opt-in Save/reload check, confirm checked consent persisted; read network/storage again.",+ "Restore consent off and save; doctor. No successful model call was needed."+ ],+ "artifacts": [+ "evidence/F16.analytics/oct16-hard-dnt-after-replay.png"+ ],+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17872,+ "title": "[Bug]: Mock-LLM E2E Tests have failed on every main run since at least 2026-09-22 (five tests fail every run)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17872",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Mock-LLM E2E CI failures and stale test selectors; test harness coverage rather than a proved Canvas user feature failure.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17871,+ "title": "[Bug]: live ACP e2e harness crashes: __EXTENSIONS_SKILLS_DIR__ is not defined in tests/e2e/live-acp/vite-node.config.mts",+ "url": "https://github.com/OpenHands/OpenHands/issues/17871",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Live ACP E2E harness compile-time constant setup; test harness startup rather than the production Canvas path.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17806,+ "title": "[Bug]: Changing LLM model without re-typing the API key wipes the stored key (auth errors on profile switch)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17806",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F10",+ "feature_ids": [+ "F10.edit"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "current-verification-blocked",+ "reason": "Current direct-profile Save reached provider preflight, which rejected the refused-loopback dummy endpoint before any profile write. The saved model remained unchanged and the stored key remained present. Key preservation after a completed Save is therefore blocked, not a reproduced failure or a fresh negative.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17806#issuecomment-6122466133",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "blocked",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 3,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "run": "export OH_VERIFY_RUN=<fresh owned run; observed run in live.run>",+ "dummy_profile": {+ "model": "openai/gpt-4o",+ "api_key": "private dummy value file",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 5,+ "temperature": 0.4+ },+ "neutral_profile": {+ "model": "openai/gpt-4o",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4,+ "api_key": "private dummy value file"+ },+ "profile_body": {+ "llm": {+ "model": "openai/gpt-4o",+ "api_key": "private dummy value file",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 5,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "launcher_preconditions": {+ "OPENAI_BASE_URL": "http://127.0.0.1:1/v1",+ "OPENAI_API_BASE": "http://127.0.0.1:1/v1"+ },+ "limits": "Two Advanced and one Basic same-provider model-only saves/switches. Basic intentionally drops base_url and therefore needs protected owned fallback routes, restart and doctor. Provider changes, Cloud and successful authentication/inference remain untested. Excluded setup mistake is recorded separately."+ },+ "steps": [+ "control-openhands onboard --skip",+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/qa-loopback-profile.json",+ "control-openhands api POST /api/profiles/QA_KEY_EDIT --write --data @/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/qa-key-edit-profile.json",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands browser goto /settings/llm",+ "control-openhands browser click 'testid=profile-row >> has-text=QA_KEY_EDIT >> testid=profile-menu-trigger'",+ "control-openhands browser click 'testid=profile-actions-menu >> testid=profile-edit'",+ "control-openhands browser click 'testid=sdk-section-advanced-toggle'",+ "control-openhands browser attr 'testid=llm-api-key-input' placeholder",+ "control-openhands browser fill 'testid=llm-custom-model-input' openai/gpt-4o-mini",+ "control-openhands browser value 'testid=base-url-input'",+ "control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 1000",+ "control-openhands browser wait-text 'Profile \"QA_KEY_EDIT\" updated' --timeout 60000",+ "control-openhands api GET /api/profiles/QA_KEY_EDIT --pick api_key_set",+ "control-openhands api GET /api/profiles/QA_KEY_EDIT --pick config.model",+ "control-openhands browser goto /conversations/49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands browser click 'testid=chat-input-llm-profile'",+ "control-openhands browser click 'testid=chat-input-llm-profile-option-QA_KEY_EDIT'",+ "control-openhands conversation send 49f5de70-fb5c-42be-a264-03ce103ae56e --prompt 'Reply with exactly pong. Do not run any tools.'",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\nrun=Path('/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks')\np=next((run/'private').rglob('49f5de70fb5c42bea26403ce103ae56e/base_state.json'))\nd=json.loads(p.read_text());llm=d['agent']['llm']\nresult={'conversation':d['id'],'model':llm.get('model'),'key_present':bool(llm.get('api_key')),'base_url':llm.get('base_url')}\nq=run/'evidence/F10.edit/key-edit-diagnostic.json';q.parent.mkdir(parents=True,exist_ok=True);q.write_text(json.dumps([result],indent=2)+'\\n')\nprint(result)\nPY",+ "control-openhands conversation pause 49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands api POST /api/profiles/QA_KEY_EDIT_2 --write --data @/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/qa-key-edit-profile.json",+ "control-openhands browser goto /settings/llm",+ "control-openhands browser click 'testid=profile-row >> has-text=QA_KEY_EDIT_2 >> testid=profile-menu-trigger'",+ "control-openhands browser click 'testid=profile-actions-menu >> testid=profile-edit'",+ "control-openhands browser click 'testid=sdk-section-advanced-toggle'",+ "control-openhands browser attr 'testid=llm-api-key-input' placeholder",+ "control-openhands browser fill 'testid=llm-custom-model-input' openai/gpt-4o-mini",+ "control-openhands browser value 'testid=base-url-input'",+ "control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 1000",+ "control-openhands api GET /api/profiles/QA_KEY_EDIT_2 --pick api_key_set",+ "control-openhands api GET /api/profiles/QA_KEY_EDIT_2 --pick config.model",+ "control-openhands browser network --filter QA_KEY_EDIT_2",+ "control-openhands browser goto /conversations/49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands browser click 'testid=chat-input-llm-profile'",+ "control-openhands browser click 'testid=chat-input-llm-profile-option-QA_KEY_EDIT_2'",+ "control-openhands conversation send 49f5de70-fb5c-42be-a264-03ce103ae56e --prompt 'Reply with exactly pong. Do not run any tools.'",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\nrun=Path('/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks');p=next((run/'private').rglob('49f5de70fb5c42bea26403ce103ae56e/base_state.json'));d=json.loads(p.read_text());llm=d['agent']['llm'];result={'conversation':d['id'],'model':llm.get('model'),'key_present':bool(llm.get('api_key')),'base_url':llm.get('base_url'),'profile':'QA_KEY_EDIT_2'};q=run/'evidence/F10.edit/key-edit-diagnostic.json';rows=json.loads(q.read_text());rows.append(result);q.write_text(json.dumps(rows,indent=2)+'\\n');print(result)\nPY",+ "control-openhands conversation pause 49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands api POST /api/profiles/QA_KEY_BASIC --write --data @/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/qa-key-edit-profile.json",+ "control-openhands browser goto /settings/llm",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\np=Path('/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/launcher-env.json');d=json.loads(p.read_text());d['OPENAI_BASE_URL']='http://127.0.0.1:1/v1';d['OPENAI_API_BASE']='http://127.0.0.1:1/v1';p.write_text(json.dumps(d));print({'owned_default_OPENAI_BASE_URL_is_loopback':True,'owned_default_OPENAI_API_BASE_is_loopback':True,'AWS_EC2_METADATA_DISABLED':d.get('AWS_EC2_METADATA_DISABLED')=='true'})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands browser reload",+ "control-openhands browser click 'testid=profile-row >> has-text=QA_KEY_BASIC >> testid=profile-menu-trigger'",+ "control-openhands browser click 'testid=profile-actions-menu >> testid=profile-edit'",+ "control-openhands browser click 'testid=sdk-section-basic-toggle'",+ "control-openhands browser attr 'testid=llm-api-key-input' placeholder",+ "control-openhands browser choose 'testid=llm-model-input' gpt-5.6",+ "control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 1000",+ "control-openhands browser wait-text 'Profile \"QA_KEY_BASIC\" updated' --timeout 60000",+ "control-openhands api GET /api/profiles/QA_KEY_BASIC --pick api_key_set",+ "control-openhands api GET /api/profiles/QA_KEY_BASIC --pick config.model",+ "control-openhands api GET /api/profiles/QA_KEY_BASIC --pick config.base_url",+ "control-openhands browser goto /conversations/49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands browser click 'testid=chat-input-llm-profile'",+ "control-openhands browser click 'testid=chat-input-llm-profile-option-QA_KEY_BASIC'",+ "control-openhands conversation send 49f5de70-fb5c-42be-a264-03ce103ae56e --prompt 'Reply with exactly pong. Do not run any tools.'",+ "python3 - <<'PY'\nimport json\nfrom pathlib import Path\nrun=Path('/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks');p=next((run/'private').rglob('49f5de70fb5c42bea26403ce103ae56e/base_state.json'));d=json.loads(p.read_text());llm=d['agent']['llm'];r={'conversation':d['id'],'profile':'QA_KEY_BASIC','view':'basic','model':llm.get('model'),'key_present':bool(llm.get('api_key')),'base_url':llm.get('base_url'),'owned_fallback_endpoint':'http://127.0.0.1:1/v1'};q=run/'evidence/F10.edit/key-edit-diagnostic.json';rows=json.loads(q.read_text());rows.append(r);q.write_text(json.dumps(rows,indent=2)+'\\n');print(r)\nPY",+ "control-openhands conversation pause 49f5de70-fb5c-42be-a264-03ce103ae56e",+ "control-openhands browser screenshot --feature F10.edit --name 17806-basic-key-preserved"+ ],+ "evidence": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/evidence/F10.edit/key-edit-diagnostic.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/evidence/F10.edit/17806-dummy-key-preserved.png",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/evidence/F10.edit/17806-basic-key-preserved.png"+ ],+ "observation": "All three actual saved/switched profiles kept the dummy key; Basic runtime base_url was null and its backend default endpoint was protected loopback.",+ "setup_mistakes": [+ "One premature Basic Save removed base_url and reached OpenAI with only the dummy key; validation rejected AuthenticationError, no profile save or inference succeeded. This setup mistake is excluded from product verdict/proof. The corrected Basic run was performed only after owned default routes, restart and doctor."+ ],+ "initial_profile_requests": {+ "file": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/qa-key-edit-profile.json",+ "names": [+ "QA_KEY_EDIT",+ "QA_KEY_EDIT_2",+ "QA_KEY_BASIC"+ ],+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "api_key": "private dummy value file",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 5,+ "temperature": 0.4+ },+ "include_secrets": true+ }+ },+ "changes": [+ {+ "name": "QA_KEY_EDIT",+ "view": "Advanced",+ "from": "openai/gpt-4o",+ "to": "openai/gpt-4o-mini"+ },+ {+ "name": "QA_KEY_EDIT_2",+ "view": "Advanced",+ "from": "openai/gpt-4o",+ "to": "openai/gpt-4o-mini"+ },+ {+ "name": "QA_KEY_BASIC",+ "view": "Basic",+ "from": "openai/gpt-4o",+ "to": "openai/gpt-5.6"+ }+ ],+ "replay_notes": [+ "Use freshly returned conversation IDs and fixture paths. Create QA_KEY_EDIT_2 only after the first edit to keep the historical row selector unambiguous.",+ "Read the second save result after its bounded 30s client deadline; captured browser network had POST 201 at 30.007s. The missed transient second toast is not proof.",+ "The excluded first Basic setup attempt and its rejected provider validation appear only in setup_mistakes, not in successful proof steps."+ ],+ "verified_owned_server_environment": {+ "OPENAI_BASE_URL": "http://127.0.0.1:1/v1",+ "OPENAI_API_BASE": "http://127.0.0.1:1/v1",+ "AWS_EC2_METADATA_DISABLED": "true",+ "confirmed_as_booleans_only": true+ },+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current Canvas local switch fetches fresh profile config and freshusageUUID; preserved-key Oct12 diagnostic remains historical. Root owns same-provider Basic/Advanced currentreplay underprotectedfallbacks."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 181+ }+ ],+ "map_note_present": true,+ "current_result": "blocked",+ "verification_status": "blocked",+ "prerequisite": "A working provider-validation route with approved credentials, so the direct Save can complete with the masked existing key left untouched.",+ "current_attempt": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "run_id": "backfill-18334-settings-root",+ "sdk": "1.55.0",+ "steps": [+ "Open QA_ROOT_EDIT in Settings > LLM",+ "Leave masked key untouched, change model and Save",+ "Observe provider validation connection error and no profile write",+ "Read saved model unchanged and api_key_set true; doctor healthy"+ ]+ },+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17803,+ "title": "[Bug]: Editing provider-connection base_url does not take effect on next conversation (stale snapshot in settings.json)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17803",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F11",+ "feature_ids": [+ "F11.edit",+ "F11.agent-uses-connection"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-row.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-actions-menu.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/delete-provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/llm-profiles-manager.tsx",+ "src/api/provider-connections-service/"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "The SDK provider-connection cascade/re-resolve commits40021e951/10aae0eeb exist on local branches but are NOT ancestors of pinnedv1.55.0; do not retire the historical BaseURL note as shipped."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "line": 83+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17801,+ "title": "[Bug]: Adding or Editing an LLM does not always take you to the Basic tab",+ "url": "https://github.com/OpenHands/OpenHands/issues/17801",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.create-prefill",+ "F10.edit"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 104+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17788,+ "title": "[Bug] Cloud attachment upload rejects valid runtime data when cached fields are empty strings",+ "url": "https://github.com/OpenHands/OpenHands/issues/17788",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.attach-files"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A Cloud sandbox whose cached runtime URL/key are blank while the fresh runtime lookup is already ready.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A Cloud sandbox whose cached runtime URL/key are blank while the fresh runtime lookup is already ready.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17785,+ "title": "[Bug]: Windows Desktop v1.24.0 - Saving DeepSeek-v4-flash LLM config throws AttributeError: 'PromptTokensDetailsWrapper' object has no attribute 'cache_creation_tokens'",+ "url": "https://github.com/OpenHands/OpenHands/issues/17785",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.create-validation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 169+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17783,+ "title": "[Bug]: Canvas App install form throws on malformed percent escapes in tree URLs",+ "url": "https://github.com/OpenHands/OpenHands/issues/17783",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F20",+ "feature_ids": [+ "F20.git-tree-url",+ "F20.install-error"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "basis": "retained-historical-note",+ "reason": "Executed both malformed ref and path tree URLs in the live install modal; each raised URI malformed, no install mutation or toast, modal remains. Reviewed local demo control installs and enables.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17774,+ "title": "[Bug]: Planner tab fails with \"Conversation file path must stay inside the workspace\" on Windows local backend",+ "url": "https://github.com/OpenHands/OpenHands/issues/17774",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.planner-empty",+ "F27.planner-plan"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "Native Windows local Agent Server and drive-letter workspace fixtures are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Native Windows local Agent Server and drive-letter workspace fixtures are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17773,+ "title": "[Bug]: Files tab always shows \"No files in workspace\" on Windows local backend",+ "url": "https://github.com/OpenHands/OpenHands/issues/17773",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-tree"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "A Windows local Agent Server using cmd.exe; this run is macOS.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A Windows local Agent Server using cmd.exe; this run is macOS.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17763,+ "title": "[Bug]: Self-hosted LOCAL_BACKEND_API_KEY is not exported to OH_SESSION_API_KEYS_0 in entrypoint.sh, causing agent-server auth bypass",+ "url": "https://github.com/OpenHands/OpenHands/issues/17763",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.docker-image"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "An owned Docker container/daemon is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "An owned Docker container/daemon is unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17710,+ "title": "[Bug]: Secret edit/deletion leads to duplicate secret being created and none being deleted",+ "url": "https://github.com/OpenHands/OpenHands/issues/17710",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F14",+ "feature_ids": [+ "F14.delete",+ "F14.edit"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F14-secrets.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/secrets-settings.tsx",+ "src/components/features/settings/secrets-settings/",+ "src/api/secrets-service.ts",+ "src/components/features/conversation/secret-request-modal.tsx",+ "src/services/secret-request.ts",+ "src/stores/secret-request-store.ts",+ "src/constants/secret-request.ts",+ "src/api/request-secret-client-tool.ts"+ ],+ "changed_canvas_source_paths": [+ "src/api/request-secret-client-tool.ts",+ "src/components/features/conversation/secret-request-modal.tsx",+ "src/components/features/settings/secrets-settings/secret-form.tsx",+ "src/constants/secret-request.ts",+ "src/services/secret-request.ts",+ "src/stores/secret-request-store.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F14-secrets.md",+ "line": 85+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17706,+ "title": "[Bug]: Canvas conversations on openhands/claude-* fail with 400 on prompt_cache_key (uvx resolves litellm>=1.99)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17706",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.default-used"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 170+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17704,+ "title": "[Bug]: A non-ASCII file name in a script bundle shows garbled (caf\u00e9.py as caf\u00c3\u00a9.py) and loses its Entrypoint badge",+ "url": "https://github.com/OpenHands/OpenHands/issues/17704",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.script-section"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Owned UTF-8 USTAR bundle caf\u00e9.py/main.py, inactive automation python3 caf\u00e9.py: actual Script header renders caf\u00c3\u00a9.py and no Entrypoint badge; ASCII control untouched.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17657,+ "title": "fix(ci): repair release ready workflow startup failures",+ "url": "https://github.com/OpenHands/OpenHands/issues/17657",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Release-ready workflow permission/startup failure; repository CI tooling.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17623,+ "title": "[Feature]: Helm chart should not automount the ServiceAccount token in the agent pod by default",+ "url": "https://github.com/OpenHands/OpenHands/issues/17623",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.helm-chart"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Helm and a disposable Kubernetes cluster are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Helm and a disposable Kubernetes cluster are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17622,+ "title": "[Bug]: Node.js version in README/SELF_HOSTING docs (22.12.x) contradicts engines.node >=24 and CI (24.15)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17622",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Node requirement documentation drift; documentation only.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17621,+ "title": "[Bug]: Compact context always fails after exactly 60s \u2014 client-side timeout aborts a condensation the server then completes",+ "url": "https://github.com/OpenHands/OpenHands/issues/17621",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A real summarizing model and conversation whose condensation takes longer than 60 seconds.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real summarizing model and conversation whose condensation takes longer than 60 seconds.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17620,+ "title": "[Bug]: Fresh-clone npm run typecheck fails with TS2307 for #/i18n/declaration (make-i18n not chained)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17620",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Fresh-clone typecheck generation order; contributor build tooling.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17619,+ "title": "[Bug]: WebSocket reconnect resubscribes from a stale `after_timestamp`, replaying the whole session",+ "url": "https://github.com/OpenHands/OpenHands/issues/17619",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F06",+ "feature_ids": [+ "F06.reconnect"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "live-not-reproduced",+ "reason": "Two current local trials retained the same open page across an owned backend restart. Each session advanced its after_seq cursor to the prior REST event count minus one, with an observed old-socket close, a new communicating connection and one visible recovery marker. Metadata only: no claim about payload replay, provider streams or Cloud.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17619#issuecomment-6122465227",+ "steps": [+ "Create an idle tools-empty dummy-loopback conversation with autotitle false; open it in the configured browser",+ "POST three genuine run:false user markers in trial1 and one in trial2; read events/count and initial socket after_seq=-1",+ "control-openhands restart; doctor; keep the browser and page open",+ "POST a recovery marker with run:false; inspect finite browser events metadata and count the visible marker",+ "Check reconnected after_seq=prior count-1 and one recovery occurrence"+ ],+ "artifacts": [+ "evidence/F06.reconnect/oct16-cursor-readback.json",+ "evidence/F06.reconnect/oct16-reconnected.png"+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": "not-reproduced",+ "disposition": "parent-fresh-recheck-pending",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "current_result": "pass",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Create an idle tools-empty dummy-loopback conversation with autotitle false; open it in the configured browser",+ "POST three genuine run:false user markers in trial1 and one in trial2; read events/count and initial socket after_seq=-1",+ "control-openhands restart; doctor; keep the browser and page open",+ "POST a recovery marker with run:false; inspect finite browser events metadata and count the visible marker",+ "Check reconnected after_seq=prior count-1 and one recovery occurrence"+ ],+ "observations": {+ "trials": [+ {+ "trial": 1,+ "events_before_restart": 7,+ "initial_after_seq": -1,+ "reconnected_after_seq": 6,+ "recovery_message_visible_count": 1,+ "observed_close_and_new_communicating_connection": true+ },+ {+ "trial": 2,+ "events_before_restart": 3,+ "initial_after_seq": -1,+ "reconnected_after_seq": 2,+ "recovery_message_visible_count": 1,+ "observed_close_and_new_communicating_connection": true+ }+ ]+ },+ "artifacts": [+ "evidence/F06.reconnect/oct16-cursor-readback.json",+ "evidence/F06.reconnect/oct16-reconnected.png"+ ],+ "no_successful_model_inference": true+ },+ "comment_id": 6122465227,+ "refresh_published_at": "2026-10-16T04:59:20Z",+ "body_sha256": "f3fa86ff74697e5fdb1361ce168edadd55d0890eb62f8b4c7bebdfbecae4e5c6",+ "run_id": "backfill-18334-settings-root",+ "needs_current_replay": false,+ "current_verdict": "not-reproduced",+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17577,+ "title": "[Bug]: Basic LLM form cannot use a model the provider catalog does not list (ready-for-dev companion to #17233)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17577",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.model-picker"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "SDK1.55/#5703 retains verified OpenRouter entries in catalog; an off-catalog custommodel blur path is a different frontend behavior, so do not retire the historical custommodel note from the catalogfix."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 106+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17567,+ "title": "[Bug]: Switching from expanded Git diff to Planner raises a Monaco disposed-model exception",+ "url": "https://github.com/OpenHands/OpenHands/issues/17567",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.diff-view-modes"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "existing-note",+ "reason": "Existing issue link and recorded failure preserved; no fresh reproduction date claimed.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "See the issue-specific prerequisite in the linked Known issue note.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17566,+ "title": "[Bug]: Terminal tab loses command history after page reload while chat retains it",+ "url": "https://github.com/OpenHands/OpenHands/issues/17566",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.terminal-history-reload"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "Existing terminal history failure reproduced 2026-10-06 is preserved; no available model credential for a new terminal command.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17564,+ "title": "[Bug]: Mobile automation dashboard actions squeeze the heading to about 52px",+ "url": "https://github.com/OpenHands/OpenHands/issues/17564",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F21",+ "feature_ids": [+ "F21.phone"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "reason": "With 26 disabled automations in the current list view, Dashboard measures 347 px wide at 390\u00d7844 and 277 px at 320\u00d7700, both 28 px tall, with zero document horizontal overflow. The reported 52 px heading squeeze does not reproduce.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17564#issuecomment-6121791800",+ "basis": "live-check",+ "issue_state": "open",+ "previous_outcome": "not-reproduced",+ "historical_basis": null,+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "pass",+ "fresh_reproduction_claim": false,+ "disposition": "current-negative-rechecked",+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "Existing user Chrome: Automate \u2192 View mode \u2192 List view. Set 390\u00d7844, then 320\u00d7700 using the viewport capability. Read h1 bounding box, scrollWidth/clientWidth and document width; screenshot both sizes."+ ],+ "artifacts": [+ "evidence/F21.phone/current-390.png",+ "evidence/F21.phone/current-390.json",+ "evidence/F21.phone/current-320.png",+ "evidence/F21.phone/current-320.json"+ ],+ "verification_status": "not-reproduced",+ "comment_id": 6121791800,+ "refresh_published_at": "2026-10-16T04:40:12Z",+ "body_sha256": "dcd13cc95622a8c02bbb65a834abff54daf77e109212786d190d1805278e5f9a"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17563,+ "title": "[Bug]: Mobile automation edit dialog clips its header and footer",+ "url": "https://github.com/OpenHands/OpenHands/issues/17563",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.phone",+ "F23.edit-close"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "live-check",+ "reason": "At 390\u00d7844, Edit opens the replacement /automations/setup?automationId= page. Save (y65.5..93.5), Close (y65.5..93.5), and Name (y148..184) are inside the viewport; there is no old 940 px modal or negative top offset. This only retires the old modal clipping report: this check does not assess the inner setup schedule-control overflow visible in the screenshot.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "pass",+ "fresh_reproduction_claim": false,+ "disposition": "current-negative-rechecked",+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "In Chrome at 390\u00d7844, Automate \u2192 first fixture Automation actions \u2192 Edit. Read Name/Save/Close bounds and dialog count, screenshot, then Close. The old detail Edit and dashboard Edit now share the setup route; this replay used dashboard Edit."+ ],+ "artifacts": [+ "evidence/F23.phone/current-edit-390.png",+ "evidence/F23.phone/current-edit-390.json"+ ],+ "verification_status": "not-reproduced",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17563#issuecomment-6208220494",+ "comment_id": 6208220494,+ "refresh_published_at": "2026-10-16T04:40:13Z",+ "body_sha256": "714eda6cff64eafc8081cc0261ba8d904c51ddacddea45caba82a27da887097e"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17559,+ "title": "[Bug]: History pagination can show both an action and its observation (ready-for-dev companion to #16947)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17559",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.load-older-history"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A real persisted history page boundary splitting an action/observation or ACP started/completed pair; dummy-key errors do not produce these tool pairs.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real persisted history page boundary splitting an action/observation or ACP started/completed pair; dummy-key errors do not produce these tool pairs.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17546,+ "title": "[Bug]: detect ports automatically instead of always using 3001",+ "url": "https://github.com/OpenHands/OpenHands/issues/17546",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.cli-port-in-use"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. It provides no automatic frontend fallback, and the ingress override does not change that internal port.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17520,+ "title": "[Bug]: Deleting the title LLM profile leaves a stale row when settings cleanup fails",+ "url": "https://github.com/OpenHands/OpenHands/issues/17520",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.delete"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 112+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17508,+ "title": "Agent Canvas update banner keeps showing 1.19.0 after npm install -g @openhands/agent-canvas@latest (installed package is 1.20.0)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17508",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F09",+ "feature_ids": [+ "F09.update-available"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F09-settings-shell.md",+ "basis": "known-issue",+ "reason": "Isolated old/new package installs with the old process still serving, or the original stale bundle/cache capture after update.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Isolated old/new package installs with the old process still serving, or the original stale bundle/cache capture after update.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17497,+ "title": "[Bug]: Corrupted conversation data breaks the entire conversation history list instead of handling individual broken conversations",+ "url": "https://github.com/OpenHands/OpenHands/issues/17497",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "known-issue",+ "reason": "The original corrupted persisted conversation or a deterministic corruption fixture after model-switch failure and fork.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original corrupted persisted conversation or a deterministic corruption fixture after model-switch failure and fork.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17494,+ "title": "[Bug]: Transcript export leaks raw i18n template for ACP tool calls without a title",+ "url": "https://github.com/OpenHands/OpenHands/issues/17494",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.export-transcript"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "An authenticated ACP session that emits a titleless tool call, then Markdown and HTML export.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated ACP session that emits a titleless tool call, then Markdown and HTML export.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17492,+ "title": "Failed browser, glob, grep, and file editor tool calls render as successful in chat",+ "url": "https://github.com/OpenHands/OpenHands/issues/17492",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working model emitting failed browser, glob, grep, planning-editor and file-editor observations with is_error=true.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model emitting failed browser, glob, grep, planning-editor and file-editor observations with is_error=true.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17490,+ "title": "Terminal commands that fail with an ordinary non-zero exit render as successful in chat",+ "url": "https://github.com/OpenHands/OpenHands/issues/17490",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working model producing a completed terminal command with nonzero exit_code and is_error=false.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model producing a completed terminal command with nonzero exit_code and is_error=false.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17476,+ "title": "Planning socket stats overwrite the main conversation's context meter",+ "url": "https://github.com/OpenHands/OpenHands/issues/17476",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A real planning sub-conversation and main conversation with distinct nonzero usage stats.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real planning sub-conversation and main conversation with distinct nonzero usage stats.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17471,+ "title": "[Bug]: Chat attachment picker silently ignores re-picking the same file after removing it",+ "url": "https://github.com/OpenHands/OpenHands/issues/17471",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.attach-files"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "live-reproduced",+ "reason": "Same-file re-pick after removing its chip leaves attachment count zero.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F05.attach-files/issue-17471-repick.png"+ ],+ "steps": [+ "fixture file --name qa-surface.txt --content \"feature-map backfill attachment\"",+ "browser goto /",+ "browser upload testid=upload-image-input <run>/evidence/_fixtures/qa-surface.txt",+ "browser click role=button[name=\"Remove qa-surface.txt\"]",+ "browser upload testid=upload-image-input <same file>",+ "browser count role=button[name=\"Remove qa-surface.txt\"] => 0"+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17470,+ "title": "Agent server crashes with SIGILL on Apple Silicon under Colima (OpenSSL ARM capability misdetection)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17470",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.docker-image"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Affected Apple Silicon Colima ARM container environment and Docker daemon are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected Apple Silicon Colima ARM container environment and Docker daemon are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17436,+ "title": "Prompt-preset automations created on agent-canvas 1.17.0 fail every run with 500 on /api/conversations/{id}/events (preset template finish-tool wiring vs. pinned agent-server 1.46.0)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17436",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F22",+ "feature_ids": [+ "F22.setup-review-confirm"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md",+ "basis": "retained-historical-note",+ "reason": "A real LLM key and the reported legacy release/template pair are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A real LLM key and the reported legacy release/template pair are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17429,+ "title": "[Bug]: 500 error with /btw",+ "url": "https://github.com/OpenHands/OpenHands/issues/17429",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.slash-btw"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "The exact pending tool/streaming state that triggers the generator-pickle error, plus a working model for the side question.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The exact pending tool/streaming state that triggers the generator-pickle error, plus a working model for the side question.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17426,+ "title": "[Bug]: Nit: agent profile says \"set as active\" instead of \"set as default\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/17426",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F13",+ "feature_ids": [+ "F13.set-active"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 81+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17391,+ "title": "[Bug]: ChatGPT Business Subscriptions",+ "url": "https://github.com/OpenHands/OpenHands/issues/17391",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.subscription"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 171+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17390,+ "title": "HELP: CLOUD",+ "url": "https://github.com/OpenHands/OpenHands/issues/17390",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.cloud-sandbox-states"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Affected authenticated Cloud account/session and token-provisioning reproduction are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected authenticated Cloud account/session and token-provisioning reproduction are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17386,+ "title": "[Bug]: Workspace grouping disappears across browsers or after clearing site data",+ "url": "https://github.com/OpenHands/OpenHands/issues/17386",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.grouped-folders"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "live-reproduced",+ "reason": "Fresh browser puts existing workspace-bound conversations into No workspace.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F04.grouped-folders/issue-17386-after-reset.png"+ ],+ "steps": [+ "Create and register a workspace, then create empty conversations from its sidebar launch-workspace entry.",+ "View > layout-preset-by-workspace; capture groups.",+ "browser reset; onboard --skip; View > layout-preset-by-workspace.",+ "browser snapshot conversation-panel => prior conversations under No workspace; registered workspace remains.",+ "conversation status <id> readback retains workspace.working_dir matching the registered folder."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17384,+ "title": "[Bug]: File-based subagents use encrypted provider API key without decrypting it",+ "url": "https://github.com/OpenHands/OpenHands/issues/17384",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.launch-child"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "A real encrypted provider connection/profile, reviewed file-based worker fixture and authorized model key are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A real encrypted provider connection/profile, reviewed file-based worker fixture and authorized model key are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17383,+ "title": "[Bug]: Browser tooling fails due to incompatible MCP dependency version",+ "url": "https://github.com/OpenHands/OpenHands/issues/17383",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.browser-screenshot"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "Actual pinned runtime browser-use 0.11.13/mcp 1.30.0/fastmcp 3.4.8 BrowserToolExecutor constructs. The report\u2019s supported Canvas/global browser operation still needs a genuine browser-enabled agent launch and authorized LLM to issue a browser_* operation; a constructor probe alone does not establish that user flow.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Actual pinned runtime browser-use 0.11.13/mcp 1.30.0/fastmcp 3.4.8 BrowserToolExecutor constructs. The report\u2019s supported Canvas/global browser operation still needs a genuine browser-enabled agent launch and authorized LLM to issue a browser_* operation; a constructor probe alone does not establish that user flow."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17371,+ "title": "[Bug]: rotated session key isn't reconciled into openhands-backends \u2192 silent permanent 401 loop",+ "url": "https://github.com/OpenHands/OpenHands/issues/17371",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.add-agent-server",+ "F25.recovery-gate"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Added QA_Rotation_Legacy for the identical origin through Add Backend using the owned session-key file. After restart --rotate-key and reload, read-only comparison shows default-local matches injected key but the manually registered active entry does not. UI shows Disconnected and Unauthorized; core unreconciled credential persists, although there is visible status feedback in this version.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17271,+ "title": "[Bug]: New chat retains context/state from the previous conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/17271",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A completed conversation with nonzero token usage before New Chat; a dummy-key run supplies no genuine context usage.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A completed conversation with nonzero token usage before New Chat; a dummy-key run supplies no genuine context usage.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17263,+ "title": "[Bug]: Context meter stays pinned after a successful compaction once the condenser has its own usage entry",+ "url": "https://github.com/OpenHands/OpenHands/issues/17263",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A real successful compaction whose condenser usage remains larger than the default agent usage.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real successful compaction whose condenser usage remains larger than the default agent usage.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17261,+ "title": "[Bug]: Cannot add LLM profile with advanced settings",+ "url": "https://github.com/OpenHands/OpenHands/issues/17261",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.custom-model",+ "F10.connection-link"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 115+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17257,+ "title": "useStartTasks returns a hardcoded empty array \u2014 in-progress start-task cards have not rendered since #278",+ "url": "https://github.com/OpenHands/OpenHands/issues/17257",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.start-task-cards"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "known-issue",+ "reason": "An authenticated Cloud backend producing an in-progress sandbox/repository start task.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Cloud backend producing an in-progress sandbox/repository start task.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17239,+ "title": "[Bug]: model picker and LLM-profile picker clipped at top of viewport",+ "url": "https://github.com/OpenHands/OpenHands/issues/17239",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.llm-profile-picker"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "October12 Local eight-profile Home/conversation viewport checks passed. Original multi-row Cloud and real ACP model pickers remain unverified; a Local profile picker is not those consumers.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-specific-cloud-prerequisite",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Approved Cloud account or real authenticated ACP session with enough model rows to reproduce clipping, at original short viewport.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17233,+ "title": "[Bug]: Openrouter custom model support",+ "url": "https://github.com/OpenHands/OpenHands/issues/17233",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.model-picker"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Verified OpenRouter catalog repair#5703 does not prove arbitrary off-catalog Basic model input now persists; old17577 companion observation stays dated."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 106+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17213,+ "title": "Rootless Podman without subuid/subgid cannot build OpenHands runtime image",+ "url": "https://github.com/OpenHands/OpenHands/issues/17213",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "reason": "The report is OpenHands 0.54.0 runtime-image construction through ToolShield/rootless Podman, including its legacy runtime Docker template; those runtime-builder paths and UI are absent from this Agent Canvas frontend/Agent Server launcher checkout.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17198,+ "title": "[Bug]: Agent can't access terminal / file system properly: [Command finished execution with no output]",+ "url": "https://github.com/OpenHands/OpenHands/issues/17198",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "Affected Windows desktop installation, project definitions and a deterministic failing command fixture are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected Windows desktop installation, project definitions and a deterministic failing command fixture are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17195,+ "title": "[Bug]: Unable to use 9Router as a custom LLM connection proxying Ollama",+ "url": "https://github.com/OpenHands/OpenHands/issues/17195",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.connection-link",+ "F10.custom-model"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 115+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17163,+ "title": "[Bug] [Website]: Access denied to model in Cloud (online) - Free account",+ "url": "https://github.com/OpenHands/OpenHands/issues/17163",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.cloud-sandbox-states"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Affected Free Cloud account, model entitlements and organization history are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected Free Cloud account, model entitlements and organization history are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17162,+ "title": "[Bug]: /launch trust prompt shows an attacker-controlled source as a GitHub repo coordinate",+ "url": "https://github.com/OpenHands/OpenHands/issues/17162",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F19",+ "feature_ids": [+ "F19.launch-trust"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-plugins.tsx",+ "src/routes/launch.tsx",+ "src/components/features/plugins/",+ "src/components/features/launch/",+ "src/utils/plugin-launch-url.ts",+ "src/hooks/mutation/use-install-plugin.ts",+ "src/hooks/mutation/use-set-plugin-enabled.ts",+ "src/hooks/mutation/use-uninstall-plugin.ts",+ "src/hooks/mutation/use-refresh-plugin.ts"+ ],+ "changed_canvas_source_paths": [],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 90+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17161,+ "title": "[Bug]: markdown sanitizer lets untrusted content set rel=\"opener\" and any target on links",+ "url": "https://github.com/OpenHands/OpenHands/issues/17161",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.markdown-rendering"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A real reasoning, activated-skill, tool-description or error-detail body rendered without includeStandard. The attempted skills dialog exposed public skills only and did not reach the affected chat renderer.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real reasoning, activated-skill, tool-description or error-detail body rendered without includeStandard. The attempted skills dialog exposed public skills only and did not reach the affected chat renderer.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17148,+ "title": "test-and-build (windows) reports green while lint, typecheck and the test suite are gated off",+ "url": "https://github.com/OpenHands/OpenHands/issues/17148",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Windows CI lane naming/test gating; repository validation workflow.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17135,+ "title": "Agent Canvas subscription profile editor falls back model and preflight skips subscription auth",+ "url": "https://github.com/OpenHands/OpenHands/issues/17135",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.subscription"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17128,+ "title": "[Bug]: automation menu is cut off and some options are unreachable",+ "url": "https://github.com/OpenHands/OpenHands/issues/17128",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F21",+ "feature_ids": [+ "F21.kebab-menu",+ "F21.phone"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "reason": "Opened the final (26th) list-row action menu at 390\u00d7844 and 1440\u00d71000. Its 160\u00d7236 menu and 150\u00d736 Delete button remain fully inside both viewports. No option is clipped in this setup.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/17128#issuecomment-6121792158",+ "basis": "live-check",+ "issue_state": "open",+ "previous_outcome": "not-reproduced",+ "historical_basis": null,+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "pass",+ "fresh_reproduction_claim": false,+ "disposition": "current-negative-rechecked",+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "Automate \u2192 List view. At 390\u00d7844 click the last Automation actions button (scrolls it into view). Read the menu and Delete bounds; close with Escape. Repeat at 1440\u00d71000."+ ],+ "artifacts": [+ "evidence/F21.kebab-menu/current-phone.png",+ "evidence/F21.kebab-menu/current-phone.json",+ "evidence/F21.kebab-menu/current-desktop.png",+ "evidence/F21.kebab-menu/current-desktop.json"+ ],+ "verification_status": "not-reproduced",+ "comment_id": 6121792158,+ "refresh_published_at": "2026-10-16T04:40:14Z",+ "body_sha256": "b4ec1cefee549ee4f87a9cd0e983ac7de40bcef02af532f190917be010efedd0"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17121,+ "title": "[Bug]: Live run phases never render against automation 1.9.0+ (current_phase)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17121",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.activity-log",+ "F23.run-status-polling"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Actual bounded local script run API current_phase=Starting automation, no structured phase fields; Activity Log reads only Running and contains no phase text.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "See the issue-specific prerequisite in the linked Known issue note."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17112,+ "title": "[Bug]: Electron windows can navigate off the loopback app (main window, redirects, nested popups)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17112",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.desktop-main-window",+ "F26.desktop-external-links"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "A runnable owned Electron shell and repeatable popup/redirect fixtures are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A runnable owned Electron shell and repeatable popup/redirect fixtures are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17108,+ "title": "[Bug]: The repository picker lists a search result for input the user has already cleared",+ "url": "https://github.com/OpenHands/OpenHands/issues/17108",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.open-repository-dialog"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "known-issue",+ "reason": "An authenticated Cloud Git provider and a real repository search response that remains pending after clearing the input.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Cloud Git provider and a real repository search response that remains pending after clearing the input.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17094,+ "title": "docs: stale openhands-dev persistence paths in DEVELOPMENT.md",+ "url": "https://github.com/OpenHands/OpenHands/issues/17094",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Development persistence path documentation drift; documentation only.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17092,+ "title": "docs: stale session-api-key.txt path in AGENTS.md and helm NOTES",+ "url": "https://github.com/OpenHands/OpenHands/issues/17092",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Session API-key filename documentation drift; documentation only.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17069,+ "title": "[Bug]: .env.sample comments disagree with local launchers",+ "url": "https://github.com/OpenHands/OpenHands/issues/17069",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": ".env.sample launcher comments; documentation only, with no runtime change requested.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17061,+ "title": "[Bug]: Renaming a conversation to whitespace only saves an empty title",+ "url": "https://github.com/OpenHands/OpenHands/issues/17061",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.rename-inline"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "live-reproduced",+ "reason": "Whitespace sends PATCH title empty; current server returns422 and restores the prior title.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F07.rename-inline/issue-17061-titled.png"+ ],+ "steps": [+ "On an empty Local conversation: browser dblclick testid=conversation-name-title; fill testid=conversation-name-input \"QA Surface titled\"; browser press Enter; browser reload.",+ "browser network --clear; repeat title edit with one space; browser press Enter.",+ "browser network --filter /api/conversations --bodies => PATCH body title empty, status422.",+ "browser text testid=conversation-name-title and reload => prior title restored."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17051,+ "title": "[Bug]: Codex work in chat but not in the issue",+ "url": "https://github.com/OpenHands/OpenHands/issues/17051",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 128+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17026,+ "title": "agent-canvas: final-reply thinking block missing for text-only turns (reasoning_content not read for MessageEvents)",+ "url": "https://github.com/OpenHands/OpenHands/issues/17026",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.thinking"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A reasoning-capable live model emitting nonempty MessageEvent.llm_message.reasoning_content on a text-only turn, followed by reload.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A reasoning-capable live model emitting nonempty MessageEvent.llm_message.reasoning_content on a text-only turn, followed by reload.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16995,+ "title": "[Bug]: Pin as Home is missing from Canvas Extension pages",+ "url": "https://github.com/OpenHands/OpenHands/issues/16995",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F20",+ "feature_ids": [+ "F20.rail-entry"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "basis": "retained-historical-note",+ "reason": "Installed and enabled reviewed local demo; hover extension sidebar entry reveals no pin control, while hover Automate reveals sidebar-pin-home-toggle-automations.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16988,+ "title": "[Bug]: mock-LLM E2E reports success when the global timeout cut the run short",+ "url": "https://github.com/OpenHands/OpenHands/issues/16988",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Mock-LLM E2E reporter success marker after global timeout; test reporting tooling.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16977,+ "title": "[Bug]: DoneMarkerReporter counts retry attempts as tests, so one flaky retry ends the mock-LLM Docker E2E run early and fails it",+ "url": "https://github.com/OpenHands/OpenHands/issues/16977",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Mock-LLM E2E reporter counts retry attempts; test reporting tooling.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16954,+ "title": "[Bug]: Scope MCP server health by backend and connection revision",+ "url": "https://github.com/OpenHands/OpenHands/issues/16954",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F17",+ "feature_ids": [+ "F17.server-health"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 104+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16951,+ "title": "[Bug]: Centralize effective configuration for the next Home conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/16951",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.composer-blocked-without-llm"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "known-issue",+ "reason": "Current source still gates a named agent on its pinned LLM while launch follows the pill. The older payload fallback symptom is superseded: per-launch LLM override preserves named profile identity/settings, as the fresh F13 check confirms. Map keeps only the unresolved readiness mismatch and the current launch contract. No new credential-presence mismatch trial is claimed.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "blocked",+ "source_refs": [+ "src/hooks/use-llm-configured.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md"+ ],+ "verification_detail": "historical-reproduction",+ "prerequisite": "Current source still gates a named agent on its pinned LLM while launch follows the pill. The older payload fallback symptom is superseded: per-launch LLM override preserves named profile identity/settings, as the fresh F13 check confirms. Map keeps only the unresolved readiness mismatch and the current launch contract. No new credential-presence mismatch trial is claimed."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16950,+ "title": "[Bug]: Scope Home launch targets and recent repositories by backend and organization",+ "url": "https://github.com/OpenHands/OpenHands/issues/16950",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.selection-after-reload"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "known-issue",+ "reason": "Two independently configured Local backends, or two authenticated Cloud backend/organization scopes, with distinct targets.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Two independently configured Local backends, or two authenticated Cloud backend/organization scopes, with distinct targets.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16949,+ "title": "[Bug]: Keep Files and Changes views source-driven after refresh",+ "url": "https://github.com/OpenHands/OpenHands/issues/16949",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-refresh"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "An authenticated Cloud repository for the refresh-query mismatch, or real file-editor observations causing an in-place M-to-D Git refetch.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Cloud repository for the refresh-query mismatch, or real file-editor observations causing an in-place M-to-D Git refetch.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16948,+ "title": "[Bug]: Preserve unsaved SDK settings edits across background refetches",+ "url": "https://github.com/OpenHands/OpenHands/issues/16948",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F15",+ "feature_ids": [+ "F15.save-dirty-state"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/condenser-settings.tsx",+ "src/routes/agent-context-settings.tsx",+ "src/routes/verification-settings.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/utils/sdk-settings-schema.ts"+ ],+ "changed_canvas_source_paths": [+ "src/utils/sdk-settings-schema.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "line": 81+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16947,+ "title": "[Bug]: Make UI event projection order-independent across history pages",+ "url": "https://github.com/OpenHands/OpenHands/issues/16947",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.load-older-history"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A real long conversation whose REST boundary separates one action from its observation; retain the separate companion #17559.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real long conversation whose REST boundary separates one action from its observation; retain the separate companion #17559.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16927,+ "title": "[Bug]: Launcher stays alive after a required service dies with EADDRINUSE",+ "url": "https://github.com/OpenHands/OpenHands/issues/16927",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Killed only the owned required static frontend after a healthy startup. The genuine launcher remains alive and authenticated /api/settings still returns 200; UI / returns 502 Bad Gateway. This exercises the required-child fatal-exit branch using SIGKILL, not an EADDRINUSE race; exact trigger is recorded. Normal stack restored afterward.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16905,+ "title": "[Bug]: Claude Code via ACP can't see built-in skills",+ "url": "https://github.com/OpenHands/OpenHands/issues/16905",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 126+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16893,+ "title": "[Bug]: Skills list is reused after switching registered backends",+ "url": "https://github.com/OpenHands/OpenHands/issues/16893",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F18",+ "feature_ids": [+ "F18.page"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-settings.tsx",+ "src/components/features/skills/",+ "src/constants/skills-docs.ts",+ "src/hooks/use-skill-enablement.ts",+ "src/utils/skill-enablement.ts",+ "src/hooks/use-launch-skill-in-chat.ts",+ "src/api/skills-service.ts",+ "src/components/features/chat/skill-install-restart-banner.tsx",+ "src/utils/skill-install-events.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/skills/skill-type-badge.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "line": 94+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16877,+ "title": "[Bug]: Untitled terminal-status conversations fast-poll every 3 seconds indefinitely \u2014 no give-up for autotitle that never lands",+ "url": "https://github.com/OpenHands/OpenHands/issues/16877",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.open-by-url"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "October12 Local idle-null-title foreground capture showed30-second polling, not reported Cloud terminal-null-title3-second polling. Current Cloud consumer remains mapped.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-specific-cloud-prerequisite",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Cloud account and an owned terminal conversation with null title to capture genuine metadata polling.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16876,+ "title": "[Bug]: Planning-agent history flag stays true forever when sub-conversations fetch fails \u2014 infinite chat loading skeleton",+ "url": "https://github.com/OpenHands/OpenHands/issues/16876",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.history-skeleton"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "A real planning sub-conversation whose batch lookup fails or yields null, with the main history available.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real planning sub-conversation whose batch lookup fails or yields null, with the main history available.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16874,+ "title": "[Bug]: Failed cloud sandbox resume is never retried \u2014 one-shot ref guard is claimed before the await and cleared on no path",+ "url": "https://github.com/OpenHands/OpenHands/issues/16874",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.cloud-sandbox-states"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "An authenticated Cloud account, paused sandbox and controlled single-request failure are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "An authenticated Cloud account, paused sandbox and controlled single-request failure are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16872,+ "title": "[Bug]: Electron main process opens arbitrary URI schemes via shell.openExternal without a protocol allowlist",+ "url": "https://github.com/OpenHands/OpenHands/issues/16872",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.desktop-external-links"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "A runnable owned Electron shell and inert OAuth authorization URI fixture are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A runnable owned Electron shell and inert OAuth authorization URI fixture are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16871,+ "title": "[Bug]: Reverse tabnabbing in conversation card's VS Code action \u2014 window.open without noopener plus scheme-agnostic URL transformer",+ "url": "https://github.com/OpenHands/OpenHands/issues/16871",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.vscode-link"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "An authorized Cloud account/sandbox with a controlled editor exposed_urls value for the current shared-transformer scheme-rejection path. The original card action is absent and the current drawer uses noopener,noreferrer; no exploit or Cloud scheme outcome was reproduced.",+ "source_evidence": [+ "src/components/features/conversation-panel/conversation-card/conversation-card-actions.tsx:131",+ "src/components/features/conversation-panel/conversation-panel.tsx:934",+ "src/hooks/query/use-unified-vscode-url.ts:132",+ "src/components/features/conversation/conversation-tabs/drawer-vscode-link.tsx:25"+ ],+ "artifacts": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F04.download/vscode-card-action-absent-16871.png",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F04.download/vscode-card-action-absent-16871.aria.txt"+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authorized Cloud account/sandbox with a controlled editor exposed_urls value for the current shared-transformer scheme-rejection path. The original card action is absent and the current drawer uses noopener,noreferrer; no exploit or Cloud scheme outcome was reproduced.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16851,+ "title": "[Bug]: wrong model is displayed when loading existing conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/16851",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.profile-identity"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "The original conversation/model-switch history and usage record showing Opus with a Fable selector; a dummy-key conversation has no usage evidence.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original conversation/model-switch history and usage record showing Opus with a Fable selector; a dummy-key conversation has no usage evidence.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16844,+ "title": "[Bug]: Workspace lists are reused after switching registered backends",+ "url": "https://github.com/OpenHands/OpenHands/issues/16844",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.workspace-dropdown-search"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "known-issue",+ "reason": "Two independent Local backend workspace registries and parent-directory listings within the cache lifetime.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Two independent Local backend workspace registries and parent-directory listings within the cache lifetime.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16842,+ "title": "[Bug]: A late close from a replaced WebSocket marks the active connection disconnected",+ "url": "https://github.com/OpenHands/OpenHands/issues/16842",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.reconnect"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A transport fixture delaying an old socket close until its replacement is open; ordinary server restart cannot select that ordering.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A transport fixture delaying an old socket close until its replacement is open; ordinary server restart cannot select that ordering.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16839,+ "title": "[Bug]: Proxy timeout listeners accumulate on keep-alive connections",+ "url": "https://github.com/OpenHands/OpenHands/issues/16839",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Twenty sequential genuine ingress authenticated /api/conversations/search requests over one keep-alive connection, allHTTP 200; the owned ingress emitted MaxListenersExceededWarning for11 timeout listeners.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16835,+ "title": "[Bug]: Agent Canvas sidebar shows \"Cache Hit: 0\" while backend correctly records nonzero cache_read_tokens",+ "url": "https://github.com/OpenHands/OpenHands/issues/16835",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A working prompt-caching model/proxy and a real conversation reporting nonzero cache_read_tokens.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working prompt-caching model/proxy and a real conversation reporting nonzero cache_read_tokens.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16802,+ "title": "[Bug]: Codeblocks in markdown \"rich\" render are ugly",+ "url": "https://github.com/OpenHands/OpenHands/issues/16802",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-rich-plain"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "live-reproduced",+ "reason": "Rich fenced code renders nested dark/gray containers and padding; layout-only inspection.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F08.files-rich-plain/issue-16802-codeblock.png"+ ],+ "steps": [+ "Arrange <workspace>/QA_rich.md containing a heading and fenced Python print(\"qa\") block.",+ "Open conversation > right-panel-toggle > conversation-tab-files > QA_rich.md.",+ "browser screenshot --feature F08.files-rich-plain; inspect rendered image => nested dark/gray code containers.",+ "Read-only computed styles => pre padding8px12px, outer prose padding24px, borders0px. Layout-only result; no provider call."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16801,+ "title": "[Bug]: Agent starts by looking in workspace, even when started with \"no workspace\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/16801",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.launch-without-workspace"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "known-issue",+ "reason": "A working Claude Opus agent and the original task/prompt that makes it explore the default workspace; the prompt is absent.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working Claude Opus agent and the original task/prompt that makes it explore the default workspace; the prompt is absent.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16774,+ "title": "[Bug]: Mac Installer Produces Damaged Binary",+ "url": "https://github.com/OpenHands/OpenHands/issues/16774",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.desktop-boot-splash",+ "F26.desktop-main-window"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "The reported signed/notarized/quarantined macOS release DMG and an owned installation fixture are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "The reported signed/notarized/quarantined macOS release DMG and an owned installation fixture are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16752,+ "title": "[Bug]: Idle conversations show up the same as actively working conversations",+ "url": "https://github.com/OpenHands/OpenHands/issues/16752",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.card-live-status"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "live-reproduced",+ "reason": "An empty IDLE conversation uses a green active dot; no pulse animation was observed.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F04.card-live-status/issue-16752-empty.png"+ ],+ "steps": [+ "llm set --profile QA_surface --model openai/gpt-4o --base-url http://127.0.0.1:9 --api-key-env QA_DUMMY_KEY --no-validate (dummy value only).",+ "fixture git-repo --name qa-surface-repo; workspace open <run>/workspace/qa-surface-repo.",+ "Sidebar folder-plus > launch-workspace qa-surface-repo, without sending a message.",+ "conversation status <id> => idle, title null.",+ "Read-only browser eval of conversation-status elements => conversation-status-active, green dot, no animation class."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16750,+ "title": "[Bug]: Conversation list does not load while automation is running",+ "url": "https://github.com/OpenHands/OpenHands/issues/16750",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "known-issue",+ "reason": "A prompt automation or native delegated task that remains running during conversation-list reload; no usable model is configured here.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A prompt automation or native delegated task that remains running during conversation-list reload; no usable model is configured here.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16735,+ "title": "[Bug]: Plugin launch payload corrupts Unicode parameter values",+ "url": "https://github.com/OpenHands/OpenHands/issues/16735",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F19",+ "feature_ids": [+ "F19.launch-unicode"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-plugins.tsx",+ "src/routes/launch.tsx",+ "src/components/features/plugins/",+ "src/components/features/launch/",+ "src/utils/plugin-launch-url.ts",+ "src/hooks/mutation/use-install-plugin.ts",+ "src/hooks/mutation/use-set-plugin-enabled.ts",+ "src/hooks/mutation/use-uninstall-plugin.ts",+ "src/hooks/mutation/use-refresh-plugin.ts"+ ],+ "changed_canvas_source_paths": [],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 92+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16727,+ "title": "[Bug]: Conversation corrupted after VM restart: \"Missing event file\"",+ "url": "https://github.com/OpenHands/OpenHands/issues/16727",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.open-by-url"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "A disposable VM interrupted at the event-write boundary or the original missing-event persisted fixture.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A disposable VM interrupted at the event-write boundary or the original missing-event persisted fixture.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16726,+ "title": "[Bug]: Missing BaseUrl for adding LLM Profile for Azure provider",+ "url": "https://github.com/OpenHands/OpenHands/issues/16726",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.model-picker"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 108+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16724,+ "title": "[Bug]: vercel-install.sh overwrites its first insteadOf rule, so the ssh:// rewrite never applies",+ "url": "https://github.com/OpenHands/OpenHands/issues/16724",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Vercel dependency-install Git rewrite configuration; deployment build tooling with no current Canvas runtime reproduction.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16720,+ "title": "[Bug]: Agent commands should run with lower priority (higher number)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16720",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "A bounded reproducible CPU-heavy workload, affected resource envelope and model key are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A bounded reproducible CPU-heavy workload, affected resource envelope and model key are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16719,+ "title": "[Bug]: Agent status always says \"thinking\" even when nothing is happening",+ "url": "https://github.com/OpenHands/OpenHands/issues/16719",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.live-activity"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "The event/status trace and deterministic stalled or errored state from the original report; its trigger is unspecified.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The event/status trace and deterministic stalled or errored state from the original report; its trigger is unspecified.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16716,+ "title": "[Bug]: Diff view is regularly out of sync",+ "url": "https://github.com/OpenHands/OpenHands/issues/16716",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "The original simultaneous-agent/shared-repository state or a deterministic stale-diff trigger; the report only suggests that timing.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original simultaneous-agent/shared-repository state or a deterministic stale-diff trigger; the report only suggests that timing.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16690,+ "title": "Agent Canvas EXTREMELY slow and unusable >3 conversations and ECONN REFUSED error",+ "url": "https://github.com/OpenHands/OpenHands/issues/16690",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.docker-image",+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Affected Windows-host Docker setup and persisted conversation/workload fixture are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected Windows-host Docker setup and persisted conversation/workload fixture are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16689,+ "title": "[Bug]: When `workspace/` paths throw a 401, call the workspace-session endpoint",+ "url": "https://github.com/OpenHands/OpenHands/issues/16689",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-load-error"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "The original fresh-VM URL, auth/cookie state and exact request sequence; the report says the trigger is unclear.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original fresh-VM URL, auth/cookie state and exact request sequence; the report says the trigger is unclear.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16686,+ "title": "[Bug]: LLM errors not propagated to the frontend",+ "url": "https://github.com/OpenHands/OpenHands/issues/16686",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A Nebius credential and its rejected DeepSeek-V3 request, or the original complete emitted error fixture.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A Nebius credential and its rejected DeepSeek-V3 request, or the original complete emitted error fixture.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16611,+ "title": "Plugin-preset automations fail with 500 on agent-server conversation/events path (prompt presets work)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16611",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F22",+ "feature_ids": [+ "F22.setup-plugin-create"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md",+ "basis": "retained-historical-note",+ "reason": "Authenticated Cloud account, model key and reviewed no-op plugin fixture are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Authenticated Cloud account, model key and reviewed no-op plugin fixture are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16601,+ "title": "[Bug]: Analytics consent reappears sporadically",+ "url": "https://github.com/OpenHands/OpenHands/issues/16601",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F01",+ "feature_ids": [+ "F01.telemetry-consent"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md",+ "basis": "known-issue",+ "reason": "The multi-day trigger and backend consent response when the modal returns; opted-in localStorage alone does not reproduce it.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The multi-day trigger and backend consent response when the modal returns; opted-in localStorage alone does not reproduce it.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16583,+ "title": "[Bug]: Readiness and PR-description checks count fenced example text as real content",+ "url": "https://github.com/OpenHands/OpenHands/issues/16583",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Issue-readiness and PR-description validation of fenced examples; repository tooling.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16582,+ "title": "use-handle-ws-events: send missing from useEffect deps \u2014 stale WebSocket closure after reconnect silently drops AgentState.PAUSED message",+ "url": "https://github.com/OpenHands/OpenHands/issues/16582",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A maximum-iteration legacy error delivered after socket replacement, with a real model run reaching that limit.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A maximum-iteration legacy error delivered after socket replacement, with a real model run reaching that limit.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16581,+ "title": "use-terminal: persistentLastCommandIndex is a module-level singleton \u2014 concurrent Terminal instances share one counter, second terminal renders no commands",+ "url": "https://github.com/OpenHands/OpenHands/issues/16581",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "A supported host fixture with simultaneous independent Terminal consumers and real command histories is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A supported host fixture with simultaneous independent Terminal consumers and real command histories is unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16578,+ "title": "[Bug]: litellm.APIConnectionError: No module named 'boto3'",+ "url": "https://github.com/OpenHands/OpenHands/issues/16578",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F10",+ "feature_ids": [+ "F10.validation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "live-not-reproduced",+ "reason": "Two genuine Canvas Bedrock starts reached the missing named-profile failure instead of a missing boto3 import. AWS metadata was disabled and both runtime/base endpoints refused loopback. This covers current Local1.55 named-profile import, not AWS credentials, successful inference or Linux fresh installs.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/16578#issuecomment-6122466700",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-reproduced",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18820,+ 18821,+ 18822,+ 18823,+ 19821+ ],+ "independent_socket_check": "all closed",+ "owned_helpers_remaining": [],+ "fixture_cleanup": "8 own LLM profiles, 2 own provider connections, own ACP agent profile, 2 dummy ACP secrets and installed plugin deleted; seeded default agent restored; dummy LLM snapshot cleared; own package/plugin sources and npm cache removed; original launcher environment/normal PATH restored. Seeded automation secret retained.",+ "evidence_files_retained": 21,+ "steps": [+ "control-openhands browser goto /settings/agents",+ "control-openhands browser click 'testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger'",+ "control-openhands browser click 'testid=agent-profile-set-active'",+ "control-openhands api DELETE /api/agent-profiles/QA_STALE_CODEX --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_API_KEY --write",+ "control-openhands api DELETE /api/settings/secrets/OPENAI_BASE_URL --write",+ "control-openhands api DELETE /api/plugins/installed/qa-resume-plugin --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_EDIT_2 --write",+ "control-openhands api DELETE /api/profiles/QA_KEY_BASIC --write",+ "control-openhands api DELETE /api/profiles/QA_QWEN --write",+ "control-openhands api DELETE /api/profiles/QA_ROUTER_MODEL --write",+ "control-openhands api DELETE /api/profiles/QA_BEDROCK --write",+ "control-openhands api DELETE /api/profiles/QA_URL_LINKED --write",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands api DELETE /api/llm/provider-connections/1b10be6572d04a32a208b501fdf77811 --write",+ "control-openhands api DELETE /api/llm/provider-connections/b326f09c63884f1282c1cad87630c999 --write",+ "control-openhands api POST /api/agent-profiles/b17c24d2-199e-4865-bb10-a78800a62475/activate --write",+ "python3 - <<'PY'\nimport shutil\nfrom pathlib import Path\nrun=Path('/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(backup,p)\nfor name in ['qa-stale-adapter','qa-resume-plugin']:\n q=run/'workspace'/name\n if q.exists():shutil.rmtree(q)\ncache=run/'private/npm-cache'\nif cache.exists():shutil.rmtree(cache)\nprint({'original_owned_launcher_environment_restored':True,'owned_package_plugin_sources_removed':True,'owned_npm_cache_removed':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands doctor",+ "control-openhands stop"+ ]+ },+ "steps": [+ "python3 - <<'PY'\nimport json,shutil\nfrom pathlib import Path\nrun=Path('/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks');p=run/'private/launcher-env.json';backup=run/'private/launcher-env.prechecks.json';shutil.copy2(p,backup);d=json.loads(p.read_text());d['AWS_EC2_METADATA_DISABLED']='true';p.write_text(json.dumps(d));print({'AWS_EC2_METADATA_DISABLED':True,'run_owned_launcher_env_backed_up':True})\nPY",+ "control-openhands restart --timeout 120",+ "control-openhands api POST /api/profiles/QA_BEDROCK --write --data @/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/qa-bedrock-profile.json",+ "control-openhands api POST /api/profiles/QA_BEDROCK/activate --write",+ "control-openhands doctor",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events f8b5b78c-3079-4049-b902-9d4a065b0ce9 --kinds ConversationErrorEvent --last 2",+ "control-openhands browser screenshot --feature F10.default-drives-conversation --name 16578-boto3-import-diagnostic",+ "control-openhands conversation start --prompt 'Reply with exactly pong. Do not run any tools.' --timeout 60",+ "control-openhands conversation events f3021128-a9cc-4d31-b6e7-a8434894871c --kinds ConversationErrorEvent --full --last 1 > '/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/bedrock-events-2.json'",+ "control-openhands conversation events f8b5b78c-3079-4049-b902-9d4a065b0ce9 --kinds ConversationErrorEvent --full --last 1 > /var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/private/bedrock-events-1.json",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands doctor"+ ],+ "evidence": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/evidence/F10.default-drives-conversation/bedrock-import-diagnostic.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-rechecks/evidence/F10.default-drives-conversation/16578-boto3-import-diagnostic.png"+ ],+ "observation": "Both starts imported boto3 and failed at missing named AWS profile before any AWS/provider request.",+ "fixture_specs": {+ "llm": {+ "model": "bedrock/anthropic.claude-3-haiku-20240307-v1:0",+ "api_key": "private dummy presence marker for Canvas gate; SDK filters this non-Bedrock key",+ "base_url": "http://127.0.0.1:1/v1",+ "aws_bedrock_runtime_endpoint": "http://127.0.0.1:1/v1",+ "aws_region_name": "us-east-1",+ "aws_profile_name": "qa_nonexistent",+ "num_retries": 0+ },+ "include_secrets": true+ },+ "setup_mistakes": [+ "Initial keyless fixture hit Canvas launch gate (llm_api_key_is_set=false), so no conversation was created; excluded from proof. Corrected dummy-only presence marker allowed the SDK Bedrock named-profile initialization."+ ],+ "uncovered_branches": [+ "valid AWS credentials/Bedrock permissions",+ "successful inference",+ "historical Linux npm dependency resolution",+ "fresh uncached dependency-resolution variants"+ ],+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "run": "Export OH_VERIFY_RUN as a fresh owned run directory; substitute actual conversation ids and private fixture paths.",+ "launcher_preconditions": {+ "AWS_EC2_METADATA_DISABLED": "true",+ "OPENAI_BASE_URL": "http://127.0.0.1:1/v1",+ "OPENAI_API_BASE": "http://127.0.0.1:1/v1"+ },+ "verification": "Set only run-private launcher environment, restart and doctor; confirm only booleans for the three flags on the process bound to the owned Agent Server port. No credentials or full environment output.",+ "profile_body": {+ "llm": {+ "model": "bedrock/anthropic.claude-3-haiku-20240307-v1:0",+ "api_key": "private dummy presence marker for Canvas gate; SDK filters this non-Bedrock key",+ "base_url": "http://127.0.0.1:1/v1",+ "aws_bedrock_runtime_endpoint": "http://127.0.0.1:1/v1",+ "aws_region_name": "us-east-1",+ "aws_profile_name": "qa_nonexistent",+ "num_retries": 0+ },+ "include_secrets": true+ },+ "scope": "Named missing profile must fail before AWS/provider HTTP; both runtime endpoints are explicitly loopback. Dummy key is only for Canvas launch gate; no actual AWS credentials."+ },+ "replay_notes": [+ "Use existing stable F10.default-used in new screenshot/evidence commands. Original artifact files remain under their captured folder path, whose first evidence tag was corrected."+ ],+ "media_review": {+ "reviewed": "all linked new screenshots visually inspected; no visible keys or credentials",+ "limits": "Bedrock screenshot shows collapsed UI error with truncated UV path/Traceback line; prefer sanitized JSON for public proof. Raw event/full stack files remain private."+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "The old boto3.Session/missingnamedprofile diagnostic is narrower than fresh uncachedLinux dependency resolution or valid AWSinference. Root owns currentSDK1.55 importreplay with metadataoff."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Arrange QA_ROOT_BEDROCK with missing qa_nonexistent AWS profile and both base/runtime endpoints127.0.0.1:1/v1",+ "Disable AWS metadata only in owned launcher and doctor",+ "Two real UI starts; wait error; inspect redacted error signatures",+ "Restore QA_ROOT_BASE"+ ],+ "observations": {+ "trials": [+ {+ "conversation": "c63021ec-56f8-4e45-b2be-daa953d0f8a5",+ "missing_boto3": false,+ "missing_named_profile": true+ },+ {+ "conversation": "2999fffa-c533-4337-8b48-92d36bacb3c6",+ "missing_boto3": false,+ "missing_named_profile": true+ }+ ]+ },+ "artifacts": [],+ "no_successful_model_inference": true+ },+ "current_result": "pass",+ "fresh_live_outcome": "not-reproduced",+ "comment_id": 6122466700,+ "refresh_published_at": "2026-10-16T04:40:15Z",+ "body_sha256": "0c037f6154fdf56c13d71961c4f477f68025777f740c09fe4dd145a485dde191",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "Arrange QA_ROOT_BEDROCK with missing qa_nonexistent AWS profile and both base/runtime endpoints127.0.0.1:1/v1",+ "Disable AWS metadata only in owned launcher and doctor",+ "Two real UI starts; wait error; inspect redacted error signatures",+ "Restore QA_ROOT_BASE"+ ],+ "artifacts": [],+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16551,+ "title": "[Bug]: [Windows] Browser tool prioritizes system Chrome over bundled Playwright Chromium, causing silent CDP hang in headless mode",+ "url": "https://github.com/OpenHands/OpenHands/issues/16551",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.browser-screenshot"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "Native Windows with both system Chrome and Playwright Chromium plus a model key is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Native Windows with both system Chrome and Playwright Chromium plus a model key is unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16530,+ "title": "IMP: Security issue is disclosed but still not fixed",+ "url": "https://github.com/OpenHands/OpenHands/issues/16530",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "reason": "This is a request to disclose an unspecified security report and request a CVE; it names no Canvas behavior, affected feature, vulnerability or reproducible failure, so there is no feature-map surface to classify.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16526,+ "title": "403 error when selecting an agent in the onboarding agent selection modal",+ "url": "https://github.com/OpenHands/OpenHands/issues/16526",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F01",+ "feature_ids": [+ "F01.onboarding-choose-agent"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md",+ "basis": "known-issue",+ "reason": "An authenticated hosted Cloud account with the reported organization-membership mismatch during onboarding.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated hosted Cloud account with the reported organization-membership mismatch during onboarding.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16511,+ "title": "[Bug]: Docker quickstart fails: bind-mount permission error kills agent-server and automation silently",+ "url": "https://github.com/OpenHands/OpenHands/issues/16511",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.docker-image"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "An owned Docker daemon/image and disposable mount ownership fixture are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "An owned Docker daemon/image and disposable mount ownership fixture are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16506,+ "title": "[Bug]: Flaky CI failure \u2014 ReferenceError: ProgressEvent is not defined",+ "url": "https://github.com/OpenHands/OpenHands/issues/16506",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "MSW/jsdom ProgressEvent test teardown race; test runner lifecycle.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16492,+ "title": "Backend API keys live in origin-readable localStorage \u2014 any same-origin script can read all of them",+ "url": "https://github.com/OpenHands/OpenHands/issues/16492",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.manage-backends",+ "F25.add-agent-server"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Read-only same-origin browser inspection records the number of registered backend entries and nonempty plaintext apiKey values; the retained probe now shows both owned entries are readable. No credential value or whole storage JSON was printed.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16455,+ "title": "[Bug]: Duplicate messages from the agent in 1.12.0",+ "url": "https://github.com/OpenHands/OpenHands/issues/16455",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.events-match"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "The original duplicated event/stream trace or deterministic prompt and timing; the report provides no reproducible trigger.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original duplicated event/stream trace or deterministic prompt and timing; the report provides no reproducible trigger.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16454,+ "title": "[Bug]: Diff list is showing duplicate entries",+ "url": "https://github.com/OpenHands/OpenHands/issues/16454",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "The original Git changes response and update sequence producing duplicate rows; the report provides no trigger.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original Git changes response and update sequence producing duplicate rows; the report provides no trigger.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16430,+ "title": "[Bug]: File upload does not have an indication of progress",+ "url": "https://github.com/OpenHands/OpenHands/issues/16430",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.attach-files"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A remote or throttled real upload that stays in flight long enough to inspect progress and composer state.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A remote or throttled real upload that stays in flight long enough to inspect progress and composer state.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16399,+ "title": "[Bug]: Pinned conversations from later pages are removed after reload",+ "url": "https://github.com/OpenHands/OpenHands/issues/16399",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.pin"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "live-reproduced",+ "reason": "A page-two pin disappears after first-page reload although the conversation still exists.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F04.pin/issue-16399-reload.png"+ ],+ "steps": [+ "Create23 empty conversations via sidebar folder-plus workspace launch, leaving the older QA Surface titled conversation outside the first20.",+ "browser goto /; browser reload; browser click role=button[name=\"Load more\"].",+ "browser hover conversation-card with QA Surface titled; click its conversation-pin-toggle control.",+ "browser text testid=conversation-panel-pinned-section => Pinned, QA Surface titled.",+ "browser reload; browser count testid=conversation-panel-pinned-section =>0; conversation still retained on server."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16383,+ "title": "Session API key passed as plaintext CLI arg to static-server.mjs (visible via ps/proc)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16383",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Inspect only the owned static-server process for frontend18813: session-key equals a literal argv value. Evidence records booleans only; no credential/argv string exposed.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16374,+ "title": "RemoteSandboxService: extend dedicated read-session fix to remaining read paths (follow-up to #14769, finding #1)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16374",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "Cloud app-server SQLAlchemy read-session atomicity has no Canvas surface or client-owned contract; implementation is outside this frontend.",+ "family": null,+ "map_file": null,+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "scope-reconfirmed",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16360,+ "title": "Agent status text overflows vertically on mobile",+ "url": "https://github.com/OpenHands/OpenHands/issues/16360",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "The issue is labeled Legacy Cloud and describes the removed v0 Cloud composer/status layout, not the current Canvas composer.",+ "family": null,+ "map_file": null,+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "scope-reconfirmed",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16356,+ "title": "Conversation LLM config is frozen at creation time \u2014 resuming after API key rotation fails permanently with LLMAuthenticationError",+ "url": "https://github.com/OpenHands/OpenHands/issues/16356",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.open-by-url"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "A working provider key that can be rotated/revoked, an existing conversation using it, and a replacement profile.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working provider key that can be rotated/revoked, an existing conversation using it, and a replacement profile.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16341,+ "title": "[Bug]: Code-explorer subagent frequently errors",+ "url": "https://github.com/OpenHands/OpenHands/issues/16341",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working model with native code-explorer delegation and the original repository task; no delegated-tool output can be generated without credentials.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model with native code-explorer delegation and the original repository task; no delegated-tool output can be generated without credentials.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16339,+ "title": "[Bug] Agent not informed of file location for files uploaded via canvas in cloud",+ "url": "https://github.com/OpenHands/OpenHands/issues/16339",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.attach-files"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "An authenticated Cloud upload sandbox and working model to inspect the delivered filename/path notification and read the uploaded file.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Cloud upload sandbox and working model to inspect the delivered filename/path notification and read the uploaded file.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16331,+ "title": "[Bug]: file preview panel stops working during streaming",+ "url": "https://github.com/OpenHands/OpenHands/issues/16331",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-tree"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "A real streaming/tool-running conversation while selecting a second workspace file.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real streaming/tool-running conversation while selecting a second workspace file.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16300,+ "title": "[Bug]: litellm build fails with rustc version mismatch on macOS/Linux",+ "url": "https://github.com/OpenHands/OpenHands/issues/16300",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "An uncached isolated reproduction of the reported LiteLLM/Rust dependency toolchain is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "An uncached isolated reproduction of the reported LiteLLM/Rust dependency toolchain is unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16298,+ "title": "[Bug]: slack-monitor skill is not available in 1.9.0",+ "url": "https://github.com/OpenHands/OpenHands/issues/16298",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F18",+ "feature_ids": [+ "F18.skill-in-chat"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-settings.tsx",+ "src/components/features/skills/",+ "src/constants/skills-docs.ts",+ "src/hooks/use-skill-enablement.ts",+ "src/utils/skill-enablement.ts",+ "src/hooks/use-launch-skill-in-chat.ts",+ "src/api/skills-service.ts",+ "src/components/features/chat/skill-install-restart-banner.tsx",+ "src/utils/skill-install-events.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/skills/skill-type-badge.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "line": 117+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16297,+ "title": "[Bug]: Default \"Local\" backend name is confusing when on a VM",+ "url": "https://github.com/OpenHands/OpenHands/issues/16297",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.selector-dropdown"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Served the unchanged UI on an owned protected non-loopback LAN origin, with auth-required and no injected key. Onboarding default Host Name is Local despite the LAN URL; after authenticating with only the private run key and skipping remaining setup, the actual selector reads Connected Local. This verifies the non-loopback naming branch, not a VM installation.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16296,+ "title": "[Bug]: Hide claude code config fields if connected",+ "url": "https://github.com/OpenHands/OpenHands/issues/16296",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F13",+ "feature_ids": [+ "F13.acp-credentials"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 127+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16277,+ "title": "Automate tab crashes on catalog entry with missing `requires` (guard getIntegrationIds)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16277",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F22",+ "feature_ids": [+ "F22.template-card",+ "F22.templates-page"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md",+ "basis": "retained-historical-note",+ "reason": "Frozen-source isolated copy with only private @openhands/extensions custom-automation manifest.requires removed; build served correctly from build/ against the healthy owned backend. Actual /automations/templates renders Application Error with TypeError: Cannot read properties of undefined (reading integrations), crashing the entire catalog instead of isolating the malformed entry. Shared source/build untouched.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16263,+ "title": "Fix LLM selector reverting to 'Select a model' for failed conversations",+ "url": "https://github.com/OpenHands/OpenHands/issues/16263",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.profile-identity"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "live-reproduced",+ "reason": "An errored conversation with an unmatched model shows Select a model despite loaded profiles.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F05.profile-identity/issue-16263-error.png"+ ],+ "steps": [+ "Create a Local conversation using dummy QA_surface profile with openai/gpt-4o and closed loopback base_url.",+ "browser reset; onboard --skip to clear browser profile stamp; reopen conversation.",+ "Send \"Reply with only QA. Do not run any tools.\" from chat-input; closed loopback later returns error.",+ "llm set same QA_surface profile to openai/gpt-4o-mini, same dummy loopback credentials, --no-validate.",+ "browser reload; browser text chat-input-llm-profile => Select a model; conversation status => error, persisted model openai/gpt-4o."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16257,+ "title": "dev launch scripts install signal handlers as import-time side effects",+ "url": "https://github.com/OpenHands/OpenHands/issues/16257",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "reason": "The issue explicitly scopes an import-time signal-handler ownership/dev-tooling refactor and states there is no user-facing feature or child-service behavior change. No observed Canvas startup/shutdown failure is supplied for the map; concrete required-service failures remain mapped separately in #16927/#18299.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16226,+ "title": "[Bug]: FIXED frontend port results in port conflict",+ "url": "https://github.com/OpenHands/OpenHands/issues/16226",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.cli-port-in-use"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. The explicit ingress override does not choose another frontend port.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16217,+ "title": "[Bug]: npm run dev can launch Uvicorn outside the automation environment",+ "url": "https://github.com/OpenHands/OpenHands/issues/16217",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "reason": "Installed real Uvicorn 0.37.0 alone in an owned Python 3.12.6 venv (no openhands module), prepended its bin only to the private launcher PATH, then restarted the real launcher. Automation 1.20.0 starts, doctor passes, and /api/automation/health returns 200. The reported external-Uvicorn ModuleNotFoundError does not occur in this exact resolver/PATH setup. Normal PATH restored, restart and doctor pass; fallback protections remain.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/16217#issuecomment-6121792551",+ "basis": "live-check",+ "issue_state": "open",+ "previous_outcome": "not-reproduced",+ "historical_basis": null,+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "pass",+ "fresh_reproduction_claim": false,+ "disposition": "current-negative-rechecked",+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "uv venv --python 3.12 <run>/private/integrations-refresh/uvicorn-only; uv pip install --offline --python <venv>/bin/python uvicorn==0.37.0. Confirm uvicorn version 0.37.0 and importlib.util.find_spec(\"openhands\") is None.",+ "Save <run>/private/launcher-env.json; prepend only <venv>/bin to its PATH. Preserve OPENAI_BASE_URL/OPENAI_API_BASE=http://127.0.0.1:1 and AWS_EC2_METADATA_DISABLED=true. control-openhands restart --timeout 30 --run <run>.",+ "control-openhands doctor --skip-ui and api GET /api/automation/health. Restore original PATH, restart --timeout 30, doctor --skip-ui."+ ],+ "artifacts": [+ "evidence/F26.runtime-services/current-standalone-uvicorn-16217.json",+ "evidence/F26.runtime-services/current-restored-doctor.json"+ ],+ "verification_status": "not-reproduced",+ "cleanup": {+ "normal_path_restored": true,+ "restored_stack_restarted": true,+ "doctor_ok": true,+ "fallback_endpoint_protection_retained": true,+ "fixture_venv_remains_private_and_inactive": true+ },+ "comment_id": 6121792551,+ "refresh_published_at": "2026-10-16T04:40:16Z",+ "body_sha256": "427b257cfce37d202f6ba57e94677467081f4ad4c6d39c1df89b348291e3719b"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16210,+ "title": "[Cloud Backends] Plugins not available when starting a new conversation",+ "url": "https://github.com/OpenHands/OpenHands/issues/16210",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.plugin-picker"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "known-issue",+ "reason": "An authenticated Cloud organization with configured account plugins before opening New Chat > Plugins.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Cloud organization with configured account plugins before opening New Chat > Plugins.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16208,+ "title": "[Bug]: Agent Canvas 1.7.2: POST /api/conversations returns 500 on Fedora Rawhide (Python 3.14, local npm install)",+ "url": "https://github.com/OpenHands/OpenHands/issues/16208",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Fedora Rawhide/Python 3.14.6 environment, reported SDK/runtime and model key are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Fedora Rawhide/Python 3.14.6 environment, reported SDK/runtime and model key are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16161,+ "title": "CI stability audit: unpinned/floating dependencies that can flip CI from green to red",+ "url": "https://github.com/OpenHands/OpenHands/issues/16161",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "CI/dependency publication stability audit; contributor build and release tooling.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16155,+ "title": "[Bug]: \"Run now\" button renders incorrectly when the automation title is long",+ "url": "https://github.com/OpenHands/OpenHands/issues/16155",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F21",+ "feature_ids": [+ "F21.list-row",+ "F21.run-now"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "reason": "The exact title \"GitHub PR Reviewer: OpenHands/software-agent-sdk for @all-hands-bot\" appears in a real disabled list row. Its current icon-only Run now button measures 32\u00d732 px with empty text, so the reported Run/now wrapping does not reproduce.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/16155#issuecomment-6121792984",+ "basis": "live-check",+ "issue_state": "open",+ "previous_outcome": "not-reproduced",+ "historical_basis": null,+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "pass",+ "fresh_reproduction_claim": false,+ "disposition": "current-negative-rechecked",+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "At 1440\u00d71000 in Automate \u2192 List view, search \"GitHub PR Reviewer\". Read the exact title and Run now dimensions; leave the disabled automation unexecuted."+ ],+ "artifacts": [+ "evidence/F21.list-row/current-long-title.png",+ "evidence/F21.list-row/current-long-title.json"+ ],+ "verification_status": "not-reproduced",+ "comment_id": 6121792984,+ "refresh_published_at": "2026-10-16T04:40:17Z",+ "body_sha256": "c05aef06ef313ae4e9972405c661db2017e49e3a5e5045dac93832d9cd07dfe2"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15946,+ "title": "Show conversations hidden by active/stale agent-server leases",+ "url": "https://github.com/OpenHands/OpenHands/issues/15946",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "known-issue",+ "reason": "A copied lease-bearing persisted conversation and second owner/expired-lease state; never alter another live run's state.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A copied lease-bearing persisted conversation and second owner/expired-lease state; never alter another live run's state.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15833,+ "title": "[Bug]: Missing text chat when Create Automation button is clicked in modal",+ "url": "https://github.com/OpenHands/OpenHands/issues/15833",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F21",+ "feature_ids": [+ "F21.add-menu",+ "F22.agent-assisted-setup"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "reason": "Add Automation \u2192 Create Automation opens /automations/setup with a New Automation editor, Name, Prompt, schedule and a visible \"Need help creating an automation?\" assistance input. The old modal-to-empty-chat branch no longer reproduces in this entry path.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/15833#issuecomment-6121793408",+ "basis": "live-check",+ "issue_state": "open",+ "previous_outcome": "not-reproduced",+ "historical_basis": null,+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "pass",+ "fresh_reproduction_claim": false,+ "disposition": "current-negative-rechecked",+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "At 1440\u00d71000 click Add Automation \u2192 Create Automation. Wait for the current New Automation setup page and inspect its form and assistance input; do not send a model prompt."+ ],+ "artifacts": [+ "evidence/F22.agent-assisted-setup/current-entry.png",+ "evidence/F22.agent-assisted-setup/current-entry.txt"+ ],+ "verification_status": "not-reproduced",+ "comment_id": 6121793408,+ "refresh_published_at": "2026-10-16T04:40:18Z",+ "body_sha256": "19661c775c488c4fdc32a067a2916ce5499d43376bfb0449afe4c114da429c77"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15822,+ "title": "Bug: Cloud-proxy calls continue after switching to local backend",+ "url": "https://github.com/OpenHands/OpenHands/issues/15822",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.switch-backend",+ "F25.health-status",+ "F25.cloud-org-rows"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "A real registered authenticated Cloud backend/account/organization is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A real registered authenticated Cloud backend/account/organization is unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15813,+ "title": "[Bug]: Condenser Max Number of Events indicator can go under 0",+ "url": "https://github.com/OpenHands/OpenHands/issues/15813",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F15",+ "feature_ids": [+ "F15.save-errors"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/condenser-settings.tsx",+ "src/routes/agent-context-settings.tsx",+ "src/routes/verification-settings.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/utils/sdk-settings-schema.ts"+ ],+ "changed_canvas_source_paths": [+ "src/utils/sdk-settings-schema.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "line": 83+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15811,+ "title": "[Bug]: Verification Settings Base/Advanced are the same",+ "url": "https://github.com/OpenHands/OpenHands/issues/15811",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F15",+ "feature_ids": [+ "F15.verification-page",+ "F15.verification-critic-fields"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/condenser-settings.tsx",+ "src/routes/agent-context-settings.tsx",+ "src/routes/verification-settings.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/utils/sdk-settings-schema.ts"+ ],+ "changed_canvas_source_paths": [+ "src/utils/sdk-settings-schema.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "line": 85+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15782,+ "title": "[Bug]: Prompt disappears when a conversation is stopped",+ "url": "https://github.com/OpenHands/OpenHands/issues/15782",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.draft-persistence"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "Clarify whether prompt means unsent composer text or the already-sent history message, and capture that state before sidebar Stop.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Clarify whether prompt means unsent composer text or the already-sent history message, and capture that state before sidebar Stop.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15778,+ "title": "LLM config accepts invalid data.",+ "url": "https://github.com/OpenHands/OpenHands/issues/15778",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.create-validation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 110+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15701,+ "title": "[Bug]: Creating Conversation toast stuck visible",+ "url": "https://github.com/OpenHands/OpenHands/issues/15701",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.launch-in-workspace"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "known-issue",+ "reason": "The original stuck creation request/outcome sequence; the report is intermittent and gives no deterministic trigger.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original stuck creation request/outcome sequence; the report is intermittent and gives no deterministic trigger.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15694,+ "title": "Bug: Enable/disable skills toggle resets to enabled state unexpectedly",+ "url": "https://github.com/OpenHands/OpenHands/issues/15694",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F18",+ "feature_ids": [+ "F18.toggle"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "basis": "live-not-reproduced",+ "reason": "The owned skill stayed disabled through two eight-second samples and reloads (four aria-checked=false readbacks). This bounded sample does not settle every intermittent timing.",+ "comment_url": "https://github.com/OpenHands/OpenHands/issues/15694#issuecomment-6121789433",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-reproduced",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands fixture skill --name qa-toggle --trigger qa-toggle-ping",+ "control-openhands browser goto /skills",+ "control-openhands browser reload",+ "control-openhands browser fill 'testid=skills-search-input' qa-toggle",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250 --observe 'testid=skill-toggle-qa-toggle' --observe-ms 8000",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands api GET /api/settings --pick misc_settings.app_preferences.disabled_skills",+ "control-openhands browser reload",+ "control-openhands browser fill 'testid=skills-search-input' qa-toggle",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250 --observe 'testid=skill-toggle-qa-toggle' --observe-ms 8000",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands api GET /api/settings --pick misc_settings.app_preferences.disabled_skills",+ "control-openhands browser reload",+ "control-openhands browser fill 'testid=skills-search-input' qa-toggle",+ "control-openhands browser attr 'testid=skill-toggle-qa-toggle' aria-checked",+ "control-openhands browser screenshot --feature F18.toggle --name 15694-disabled-persists",+ "control-openhands browser click 'testid=skill-toggle-qa-toggle' --hover-first 250",+ "control-openhands doctor"+ ],+ "evidence": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker/evidence/F18.toggle/15694-disabled-persists.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "The disabled skill stayed off in two eight-second samples and reloads; the server retained its disabled name. The report is intermittent, so longer or different timing remains unverified.",+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations."+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-settings.tsx",+ "src/components/features/skills/",+ "src/constants/skills-docs.ts",+ "src/hooks/use-skill-enablement.ts",+ "src/utils/skill-enablement.ts",+ "src/hooks/use-launch-skill-in-chat.ts",+ "src/api/skills-service.ts",+ "src/components/features/chat/skill-install-restart-banner.tsx",+ "src/utils/skill-install-events.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/skills/skill-type-badge.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Intermittent toggle behavior cannot be disproved by a dated8second sample; root owns fresh bounded current sample."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "control-openhands browser goto /skills",+ "control-openhands browser click testid=skill-toggle-qa-root-legacy",+ "Wait8s; read aria-checked; reload; read aria-checked; repeat"+ ],+ "observations": {+ "aria_checked": [+ "false",+ "false",+ "false",+ "false"+ ]+ },+ "artifacts": [],+ "no_successful_model_inference": true+ },+ "current_result": "pass",+ "fresh_live_outcome": "not-reproduced",+ "comment_id": 6121789433,+ "refresh_published_at": "2026-10-16T04:40:19Z",+ "body_sha256": "bdbf48b32792e6d578e3dd3992013569f26c28c5a25b6e05fe11f9c398147ded",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "control-openhands browser goto /skills",+ "control-openhands browser click testid=skill-toggle-qa-root-legacy",+ "Wait8s; read aria-checked; reload; read aria-checked; repeat"+ ],+ "artifacts": [],+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15692,+ "title": "[Bug]: Diff still shows after agent has committed and pushed",+ "url": "https://github.com/OpenHands/OpenHands/issues/15692",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "A real agent commit/push terminal observation while the diff remains mounted, plus a disposable Git remote.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real agent commit/push terminal observation while the diff remains mounted, plus a disposable Git remote.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15676,+ "title": "[Bug]: Refreshing stale Cloud conversation on Vercel fails",+ "url": "https://github.com/OpenHands/OpenHands/issues/15676",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.cloud-sandbox-states"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Affected Vercel Cloud session, aged sandbox and exact error/deployment revision are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected Vercel Cloud session, aged sandbox and exact error/deployment revision are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15662,+ "title": "[Bug]: Cannot select existing conversation from collapsed side nav bar",+ "url": "https://github.com/OpenHands/OpenHands/issues/15662",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F02",+ "feature_ids": [+ "F02.sidebar-collapse"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F02-app-shell.md",+ "basis": "live-reproduced",+ "reason": "Collapsed rail has no existing-conversation chooser; expansion is required.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F02.sidebar-collapse/issue-15662-collapsed.png"+ ],+ "steps": [+ "Arrange an existing empty Local conversation via workspace open and sidebar folder-plus > launch-workspace.",+ "browser click testid=sidebar-collapse-toggle",+ "browser snapshot => Search commands, New Chat, Customize, Automate, Settings, Manage Backends; no conversation chooser.",+ "browser count testid=conversation-panel => 0.",+ "Restore: browser hover role=link[name=\"OpenHands Logo\"], then click testid=sidebar-collapse-toggle."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15626,+ "title": "[Bug]: Conversation can fail with LiteLLM auth error when database server at 127.0.0.1:5432 is unreachable",+ "url": "https://github.com/OpenHands/OpenHands/issues/15626",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "The original LiteLLM proxy/database-failure event or a disposable proxy with its PostgreSQL dependency unavailable.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original LiteLLM proxy/database-failure event or a disposable proxy with its PostgreSQL dependency unavailable.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15606,+ "title": "[Bug]: When re-loading ACP conversations, messages disappear or move",+ "url": "https://github.com/OpenHands/OpenHands/issues/15606",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.reload-mid-run"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "An authenticated Codex ACP session emitting intermediate messages while another window reloads history.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Codex ACP session emitting intermediate messages while another window reloads history.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15599,+ "title": "Agent creates redundant refs when working with PRs instead of using the PR source branch",+ "url": "https://github.com/OpenHands/OpenHands/issues/15599",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "The report concerns an agent choosing a Git ref strategy from a free-form task; it identifies no Canvas state, request or Git-control defect.",+ "family": null,+ "map_file": null,+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "scope-reconfirmed",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15567,+ "title": "Workspace file preview returns 401 Unauthorized on non-HTTPS non-loopback hosts",+ "url": "https://github.com/OpenHands/OpenHands/issues/15567",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-load-error"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "A plain-HTTP non-loopback deployment and browser accessing that origin, with workspace-cookie observation.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A plain-HTTP non-loopback deployment and browser accessing that origin, with workspace-cookie observation.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15543,+ "title": "[Bug]: Bad parsing for branch retrieval",+ "url": "https://github.com/OpenHands/OpenHands/issues/15543",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.git-control-bar"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "The original Git probe stdout/interleaving that temporarily puts a filename in the branch badge; dirty state alone is not a deterministic trigger.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original Git probe stdout/interleaving that temporarily puts a filename in the branch badge; dirty state alone is not a deterministic trigger.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15527,+ "title": "[Bug]: Canvas briefly shows \"disconnected\" when coming back to an idle tab",+ "url": "https://github.com/OpenHands/OpenHands/issues/15527",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.reconnect"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "The original custom-domain deployment, idle duration and tab-suspension/network transition; those details are missing.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original custom-domain deployment, idle duration and tab-suspension/network transition; those details are missing.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15519,+ "title": "[Bug]: \"files\" tab in Files side panel throws 401 Unauthorized",+ "url": "https://github.com/OpenHands/OpenHands/issues/15519",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-load-error"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "The reported Linux Docker deployment/auth topology; this isolated run uses the macOS source launcher.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The reported Linux Docker deployment/auth topology; this isolated run uses the macOS source launcher.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15509,+ "title": "[Bug]: Automation fails on Windows local-mode: \"Path must be absolute\" during tarball upload",+ "url": "https://github.com/OpenHands/OpenHands/issues/15509",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F22",+ "feature_ids": [+ "F22.setup-tarball-create"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md",+ "basis": "retained-historical-note",+ "reason": "Native Windows stack and controlled event/webhook delivery fixture are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Native Windows stack and controlled event/webhook delivery fixture are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15501,+ "title": "[Bug]: allow-scripts inside iframe",+ "url": "https://github.com/OpenHands/OpenHands/issues/15501",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-rich-plain"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "The original preview panel and file URL or minimal page identifying the intended surface; Files Rich deliberately blocks scripts, while Browser is separate.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "The original preview panel and file URL or minimal page identifying the intended surface; Files Rich deliberately blocks scripts, while Browser is separate.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15495,+ "title": "[Bug]: Bedrock documentation is lacking",+ "url": "https://github.com/OpenHands/OpenHands/issues/15495",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "historical-only",+ "reason": "The request is for missing AWS Bedrock setup documentation at docs.openhands.dev, with no concrete Canvas control or failing UI behavior. F10 consumes profile forms, not the external documentation authoring surface.",+ "family": null,+ "map_file": null,+ "issue_state": "open",+ "previous_state": "out-of-scope",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15465,+ "title": "[Bug]: Event based Automation doesn't seem to work",+ "url": "https://github.com/OpenHands/OpenHands/issues/15465",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.configuration-event",+ "F23.activity-log"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Externally reachable responder, authorized webhook/release repository and controlled real event delivery are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Externally reachable responder, authorized webhook/release repository and controlled real event delivery are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15458,+ "title": "[Bug]: Customize > Skills is showing agent.md files inside of conversation worktrees from automation-runs",+ "url": "https://github.com/OpenHands/OpenHands/issues/15458",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F18",+ "feature_ids": [+ "F18.page",+ "F18.project-skill"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-settings.tsx",+ "src/components/features/skills/",+ "src/constants/skills-docs.ts",+ "src/hooks/use-skill-enablement.ts",+ "src/utils/skill-enablement.ts",+ "src/hooks/use-launch-skill-in-chat.ts",+ "src/api/skills-service.ts",+ "src/components/features/chat/skill-install-restart-banner.tsx",+ "src/utils/skill-install-events.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/skills/skill-type-badge.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "line": 118+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15434,+ "title": "Embedded VS Code button unreachable in self-hosted mode \u2014 advertised URL uses page origin while the server binds a separate port, with no config to reconcile them",+ "url": "https://github.com/OpenHands/OpenHands/issues/15434",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.remote-backend"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Affected non-localhost HTTPS reverse proxy deployment and functioning editor binary are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected non-localhost HTTPS reverse proxy deployment and functioning editor binary are unavailable."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15430,+ "title": "OAuth MCP callback fails in Docker: localhost redirect port not exposed",+ "url": "https://github.com/OpenHands/OpenHands/issues/15430",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F17",+ "feature_ids": [+ "F17.install-oauth"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 139+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15421,+ "title": "[Bug]: Configured MCP servers not being picked up by the agent",+ "url": "https://github.com/OpenHands/OpenHands/issues/15421",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F17",+ "feature_ids": [+ "F17.agent-uses-server"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 138+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/automation",+ "number": 586,+ "title": "[Bug]: Git Sync tarball comparison depends on gzip timestamp",+ "url": "https://github.com/OpenHands/automation/issues/586",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F24",+ "feature_ids": [+ "F24.sync-now"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F24-git-sync.md",+ "basis": "known-issue",+ "reason": "Automation 1.20 canonical_tarball_bytes still uses tarfile w:gz and compares compressed bytes. Requires a real disposable Git Sync script automation, YAML-only change across different rebuild seconds, and unchanged saved upload identity readback. Serializer-only or decompressed-content equality is not current UI proof.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Automation 1.20 canonical_tarball_bytes still uses tarfile w:gz and compares compressed bytes. Requires a real disposable Git Sync script automation, YAML-only change across different rebuild seconds, and unchanged saved upload identity readback. Serializer-only or decompressed-content equality is not current UI proof.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/automation",+ "number": 565,+ "title": "[Bug]: 502 instead of 403 when the credential is not authorized for the X-Org-Id organization",+ "url": "https://github.com/OpenHands/automation/issues/565",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F25",+ "feature_ids": [+ "F25.cloud-org-rows",+ "F25.cloud-org-suspended"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "retained-historical-note",+ "reason": "Authenticated Cloud account/org-bound key and unauthorized-org fixture are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Authenticated Cloud account/org-bound key and unauthorized-org fixture are unavailable."+ },+ {+ "repository": "OpenHands/automation",+ "number": 557,+ "title": "Prompt and plugin presets ignore the selected agent profile at run time",+ "url": "https://github.com/OpenHands/automation/issues/557",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.configuration",+ "F23.run-linked-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Saved bare qa-backfill-bare agent profile (tools=[], unique persona, closed-loopback qa-offline LLM), created prompt preset then PATCHed its selected agent_profile_id. Real UI Run now created a linked conversation whose launched_agent_profile is null, selected persona is absent and default tools are present; the selected profile was ignored before the deliberate closed-loopback LLM connection failure.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/automation",+ "number": 514,+ "title": "fix(local): stop BashCommand when an automation run is cancelled",+ "url": "https://github.com/OpenHands/automation/issues/514",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F03",+ "feature_ids": [+ "F03.cancel-in-flight-run",+ "F23.activity-log",+ "F23.run-logs-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "retained-historical-note",+ "reason": "Clicked Home owned worker Cancel run; toast Automation run cancelled and Activity Log CANCELLED. The python3 worker.py remains in this run launcher ancestry after cancellation, despite no provider or external worker calls.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/automation",+ "number": 472,+ "title": "Release conversation subjects when runs are cancelled",+ "url": "https://github.com/OpenHands/automation/issues/472",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Automation subject ownership and cancellation retry contract; Canvas does not submit subject-turn idempotency keys.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/automation",+ "number": 470,+ "title": "fix: retry unsuccessful idempotent subject turns",+ "url": "https://github.com/OpenHands/automation/issues/470",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Automation subject-turn retry/idempotency API; Canvas does not submit subject turns or idempotency keys in its UI.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/automation",+ "number": 418,+ "title": "[pr-review-bot] Improvements and Bug Fixes for the pr-review-bot",+ "url": "https://github.com/OpenHands/automation/issues/418",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "PR-review-bot merge gating, review attribution and operational audit; no concrete mapped Canvas UI behavior.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/automation",+ "number": 347,+ "title": "Run summaries are only readable one automation at a time, and the SQLite engine's undeclared pool turns the resulting fan-out into 30s stalls + HTTP 500",+ "url": "https://github.com/OpenHands/automation/issues/347",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F21",+ "feature_ids": [+ "F21.card-grid",+ "F21.overview-tiles"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "basis": "retained-historical-note",+ "reason": "The owned 25-automation local dashboard completed run summaries with HTTP 200 and no 30 s stall on automation 1.19.3. This narrow live probe does not account for the issue\u2019s missing batch-summary API/pool configuration contracts or withheld original lock workload. A failing dashboard recipe needs the original DB load/lock setup and redacted service-log/HAR timing evidence, linked to Canvas #16559.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "The owned 25-automation local dashboard completed run summaries with HTTP 200 and no 30 s stall on automation 1.19.3. This narrow live probe does not account for the issue\u2019s missing batch-summary API/pool configuration contracts or withheld original lock workload. A failing dashboard recipe needs the original DB load/lock setup and redacted service-log/HAR timing evidence, linked to Canvas #16559."+ },+ {+ "repository": "OpenHands/automation",+ "number": 343,+ "title": "setup_script_path is validated, stored and tested \u2014 but never read; only a root setup.sh ever runs",+ "url": "https://github.com/OpenHands/automation/issues/343",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F22.setup-tarball-create",+ "F23.script-section",+ "F23.run-logs-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Uploaded owned tarball scripts/setup.sh writes .qa-setup-marker, main.py requires marker; persisted setup_script_path scripts/setup.sh. UI Run now genuine BashOutput exits1 with QA_NESTED_SETUP_WAS_SKIPPED, proving nested setup ignored.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/automation",+ "number": 286,+ "title": "Background scheduler, dispatcher and watchdog start once per uvicorn worker instead of once per deployment",+ "url": "https://github.com/OpenHands/automation/issues/286",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Owned multi-worker/replica automation deployment and authenticated Cloud sandbox verification context are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Owned multi-worker/replica automation deployment and authenticated Cloud sandbox verification context are unavailable."+ },+ {+ "repository": "OpenHands/automation",+ "number": 246,+ "title": "automation",+ "url": "https://github.com/OpenHands/automation/issues/246",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F21",+ "feature_ids": [+ "F21.backend-unavailable",+ "F21.list-error"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "basis": "retained-historical-note",+ "reason": "The exact 404 URL, user action, version and local launcher/service configuration are missing from the report.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "The exact 404 URL, user action, version and local launcher/service configuration are missing from the report."+ },+ {+ "repository": "OpenHands/automation",+ "number": 219,+ "title": "Automation sandbox creation can evict a user's own live session; the graceful-skip path is dead",+ "url": "https://github.com/OpenHands/automation/issues/219",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Owned disposable Cloud account at its sandbox cap with an interactive fixture is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Owned disposable Cloud account at its sandbox cap with an interactive fixture is unavailable."+ },+ {+ "repository": "OpenHands/automation",+ "number": 200,+ "title": "[post-release] Harden KV request models (extra=forbid) + fix missing-key inconsistency + version in write responses",+ "url": "https://github.com/OpenHands/automation/issues/200",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Automation KV REST request strictness and response contracts; no Canvas KV editor or consumer of these writes.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/automation",+ "number": 113,+ "title": "Bug: UI says conversations not started, when the conversation did start",+ "url": "https://github.com/OpenHands/automation/issues/113",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.run-linked-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Affected automation/run/conversation ids, deployment version and genuine plugin-run credentials are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected automation/run/conversation ids, deployment version and genuine plugin-run credentials are unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5995,+ "title": "[Bug]: Claude Code ACP cannot run Claude Opus 5.5",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5995",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation",+ "F13.acp-model-discovery"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "current-source-awaiting-live",+ "reason": "Current SDK1.55 retains Claude ACP0.63.0, the version named by the report. Canvas uses SDK registry presets and live discovery, so first-turn Opus5.5 rejection is an actual consumer. No authenticated Claude turn was attempted.",+ "source_refs": [+ "src/constants/acp-providers.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "SDKv1.55.0:openhands-sdk/openhands/sdk/settings/acp_install_catalog.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "Approved Claude credentials/session entitled to claude-opus-5-5 and a real completed turn with the pinned adapter.",+ "map_note_status": "linked Known issue with exact prerequisite",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 130+ }+ ],+ "map_note_present": true,+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5985,+ "title": "[Bug]: A list call during Update (force reinstall) drops the extension's enabled state",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5985",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F19",+ "feature_ids": [+ "F19.refresh",+ "F19.enable-toggle"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "current-source-awaiting-live",+ "reason": "Canvas Update calls the shared plugin refresh endpoint and listing can run concurrently. Disabled-state/source loss during directory replacement is therefore relevant; the issue forces copytree scheduling in Python, which is not live Canvas proof.",+ "source_refs": [+ "src/api/plugins-management-service.ts",+ "src/hooks/mutation/use-refresh-plugin.ts",+ "SDKv1.55.0:openhands-sdk/openhands/sdk/extensions/installation/manager.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "An owned disabled local plugin plus real concurrent installed-list calls overlapping its force-reinstall directory-swap window; do not patch copytree to count it as live.",+ "map_note_status": "linked Known issue with exact prerequisite",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 136+ }+ ],+ "map_note_present": true,+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5974,+ "title": "[Bug]: Workspace registry lock contention blocks /alive and the server event loop",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5974",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "known-issue",+ "reason": "An owned workspace-registry lock plus a waiting workspace mutation can block server liveness and Canvas. Current async workspaces routes call blocking store.update. Requires a bounded lock-holder with ownership/wait confirmation, concurrent liveness probes, release and successful readback; the original exact waiter observation uses Linux /proc/locks.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "An owned workspace-registry lock plus a waiting workspace mutation can block server liveness and Canvas. Current async workspaces routes call blocking store.update. Requires a bounded lock-holder with ownership/wait confirmation, concurrent liveness probes, release and successful readback; the original exact waiter observation uses Linux /proc/locks.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5969,+ "title": "[Bug]: A slow webhook receiver makes POST /run answer 409 \"already running\" on a finished conversation",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5969",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "basis": "current-source-review",+ "reason": "Canvas Run/Play and send run:true consume the lifecycle that can report finished yet reject Run or delay follow-up while webhook is pending.",+ "source_refs": [+ "src/hooks/mutation/conversation-mutation-utils.ts",+ "src/contexts/conversation-websocket-context.tsx"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Genuine successfully finished model run on owned backend with slow webhook; submit/Play before receiver returns. No provider inference authorized.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5963,+ "title": "[Bug]: RemoteConversation.state.events keeps every StreamingDeltaEvent frame with a streaming LLM",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5963",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Automation presets instantiate Python RemoteConversation but do not enumerate its state.events or compare that cache count with the server: they use their received_events callback list. Canvas uses TypeScript event stores. This report specifies extra transient deltas in Python state.events; no concrete affected Canvas widget/run outcome for that exact cache contract was identified.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5962,+ "title": "[Bug]: A continued OpenAI-gateway conversation that stops in waiting_for_confirmation (or paused) answers 504 after 120 s instead of 409",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5962",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "OpenAI-compatible /v1/chat/completions continued-conversation gateway status handling. Canvas creates/sends conversations through ConversationClient, not the OpenAI gateway.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5961,+ "title": "[Bug]: Telemetry conversation_finished always reports total_tokens_bucket and cost_bucket \"unknown\" for real runs",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5961",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Agent Server PostHog outcome token/cost bucket instrumentation. Canvas renders conversation metrics and has its own telemetry owner; it does not consume these backend analytics payloads.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5960,+ "title": "[Bug]: TelemetrySpec.salt description promises a per-process random fallback, but conversation refs are keyed by OH_SECRET_KEY and linkable across restarts",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5960",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Agent Server TelemetrySpec.salt OpenAPI description/fallback documentation; Canvas exposes consent but no TelemetrySpec salt field. No user-facing Canvas contract change in this report.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5959,+ "title": "[Bug]: Conversation(token_callbacks=[...]) with a RemoteWorkspace silently ignores the callbacks",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5959",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Python Conversation(token_callbacks=...) forwarding into RemoteConversation. Canvas receives tokens over the TypeScript WebSocket/event path and does not instantiate this Python callback API.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5952,+ "title": "[Bug]: Webhook events published during an in-flight batch POST wait for flush_delay although the buffer is full",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5952",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Agent Server outbound event WebhookSpec batch/flush scheduling, not Canvas automation inbound custom-webhook registration. No mapped Canvas setting/consumer for this receiver batching contract.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5947,+ "title": "[Bug]: TypeScript GitChange.status is typed 'added|modified|deleted|renamed' but the server sends 'ADDED|UPDATED|DELETED|MOVED'",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5947",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Canvas own status mapper accepts uppercase wire and lowercase statuses for Local and Cloud, not erroneous exported GitChange type comparisons.",+ "source_refs": [+ "src/api/git-service/agent-server-git-service.api.ts",+ "src/utils/git-status-mapper.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5946,+ "title": "[Bug]: TypeScript updateConversation is typed to return ConversationInfo but resolves with {success: true}",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5946",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Canvas ignores PATCH result then batch-fetches canonical conversation. It does not read id/title from mis-typed update response.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts:updateConversationTitle/updateConversationTags"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5941,+ "title": "[Bug]: Overlapping /v1/chat/completions requests on one continued conversation: the first request returns the second request's answer",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5941",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "OpenAI-compatible /v1/chat/completions concurrent continued requests. Canvas uses ConversationClient send/run endpoints, not this gateway.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5940,+ "title": "[Bug]: Conversation whose agent names a tool from a module the server cannot import is created, then every message is a 500",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5940",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Requires Python client-local custom tool registration and tool_module_qualnames import mismatch. Canvas encrypted-settings/profile creation does not submit that module registration contract.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5939,+ "title": "[Bug]: Telemetry from an initialized deferred (warm-pool) pod always reports deferred_init: false",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5939",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Agent Server analytics deferred_init property on warm-pool pods. Canvas does not render or consume this backend telemetry field.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5938,+ "title": "[Bug]: RemoteConversation callbacks never receive events the server recorded while the events WebSocket was reconnecting",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5938",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Automation1.20 prompt/plugin sdk_main callbacks append received events and publish ActionEvent.summary as live phase. A real socket reconnect can omit actions recorded during the gap; completion metadata loss is not established. Prerequisite: an owned remote continuation, genuine tool events across a controlled reconnect and the phase consumer visible; separate current phase defect17121 may mask this symptom.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling",+ "F23.run-logs-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "known-issue",+ "prerequisite": "Automation1.20 prompt/plugin sdk_main callbacks append received events and publish ActionEvent.summary as live phase. A real socket reconnect can omit actions recorded during the gap; completion metadata loss is not established. Prerequisite: an owned remote continuation, genuine tool events across a controlled reconnect and the phase consumer visible; separate current phase defect17121 may mask this symptom.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5937,+ "title": "[Bug]: RemoteConversation.run() on a running conversation raises ReadTimeout when the current tool call lasts over 30 s",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5937",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "Automation1.20 prompt/plugin presets attach AUTOMATION_CONVERSATION_ID, send the user prompt and call synchronous RemoteConversation.run(). Continuing an owned persisted conversation already running a genuine tool longer than30s can hit the SDK trigger read timeout while the run is healthy. Prerequisite: controlled continued automation plus its real long-running model/tool turn; a stopped dummy provider is not this join scenario.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling",+ "F23.run-logs-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "known-issue",+ "prerequisite": "Automation1.20 prompt/plugin presets attach AUTOMATION_CONVERSATION_ID, send the user prompt and call synchronous RemoteConversation.run(). Continuing an owned persisted conversation already running a genuine tool longer than30s can hit the SDK trigger read timeout while the run is healthy. Prerequisite: controlled continued automation plus its real long-running model/tool turn; a stopped dummy provider is not this join scenario.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5936,+ "title": "[Bug]: WebhookSpec descriptions give the wrong events path and say the flush timer resets on each event",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5936",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "WebhookSpec description of outbound webhook event path/flush timer. Documentation-only backend contract, distinct from Canvas automation custom inbound webhooks.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5935,+ "title": "[Bug]: OH_{FIELD}_IS_NONE=1 cannot null a value set in the agent-server config file",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5935",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F20",+ "feature_ids": [+ "F20.app-backend-view"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "basis": "known-issue",+ "reason": "Canvas consumes app_backend_ingress_url/canvas_app_backend_bridge_v1; an environment null does not clear a JSON-config public URL. Recheck an admitted bridge-capable server with those conflicting sources and the actual app-backend page; default launcher configuration does not arrange this state.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Canvas consumes app_backend_ingress_url/canvas_app_backend_bridge_v1; an environment null does not clear a JSON-config public URL. Recheck an admitted bridge-capable server with those conflicting sources and the actual app-backend page; default launcher configuration does not arrange this state.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5934,+ "title": "[Bug]: A 429 (run limit) on a continued OpenAI-gateway conversation still saves the user message, so each retry appends a duplicate",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5934",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "reason": "OpenAI-compatible /v1/chat/completions continued-conversation retry persistence. Canvas sends user messages through ConversationClient rather than this gateway.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/event-service/event-service.api.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/services/telemetry.ts"+ ],+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5922,+ "title": "[Bug]: POST /api/meta-profiles/{name}/activate with an unreadable settings.json answers an unmapped 500 naming the settings path (expected 409)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5922",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F12",+ "feature_ids": [+ "F12.activate"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F12-model-router.md",+ "basis": "live-reproduced",+ "reason": "Two real Model Router Set active attempts over owned corrupt settings showed an error toast; activation returned500 with settings-load/path detail instead of409. Corrupt bytes were not overwritten, original bytes restored in finally, doctor healthy. No model call.",+ "source_refs": [+ "src/api/meta-profiles-service/meta-profiles-service.api.ts",+ "src/components/features/settings/meta-llm-profiles/meta-llm-settings-view.tsx",+ "SDKv1.55.0:openhands-agent-server/openhands/agent_server/meta_profiles_router.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "map_note_status": "current Known failure; executable block reproduced twice",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ [+ "control-openhands",+ "api",+ "POST",+ "/api/meta-profiles/QA_CORRUPT_ROUTER",+ "--data",+ "@/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/private/root-replays/router5922.json",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/settings/meta-llm"+ ],+ [+ "control-openhands",+ "browser",+ "wait",+ "testid=meta-profile-row-QA_CORRUPT_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-menu-trigger-QA_CORRUPT_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-set-active"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/meta-profiles/QA_CORRUPT_ROUTER/activate",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--history"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-menu-trigger-QA_CORRUPT_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-set-active"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/meta-profiles/QA_CORRUPT_ROUTER/activate",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--history"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/meta-profiles/QA_CORRUPT_ROUTER",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ]+ ],+ "observations": {+ "attempts": [+ {+ "attempt": 1,+ "activation_status": 500,+ "exception_is_settings_load": true,+ "settings_path_disclosed": true,+ "corrupt_bytes_unchanged": true,+ "ui_toasts_present": true+ },+ {+ "attempt": 2,+ "activation_status": 500,+ "exception_is_settings_load": true,+ "settings_path_disclosed": true,+ "corrupt_bytes_unchanged": true,+ "ui_toasts_present": true+ }+ ],+ "settings_restored": true,+ "raw_exception_private": true+ },+ "artifacts": [+ "evidence/F12/5922-current.json"+ ],+ "no_successful_model_inference": true+ },+ "current_result": "fail",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F12-model-router.md",+ "line": 102+ }+ ],+ "map_note_present": true,+ "fresh_live_outcome": "mapped",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ [+ "control-openhands",+ "api",+ "POST",+ "/api/meta-profiles/QA_CORRUPT_ROUTER",+ "--data",+ "@/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/private/root-replays/router5922.json",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/settings/meta-llm"+ ],+ [+ "control-openhands",+ "browser",+ "wait",+ "testid=meta-profile-row-QA_CORRUPT_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-menu-trigger-QA_CORRUPT_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-set-active"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/meta-profiles/QA_CORRUPT_ROUTER/activate",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--history"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-menu-trigger-QA_CORRUPT_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-set-active"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/meta-profiles/QA_CORRUPT_ROUTER/activate",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--history"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/meta-profiles/QA_CORRUPT_ROUTER",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ]+ ],+ "artifacts": [+ "evidence/F12/5922-current.json"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced",+ "reproduction_setup_and_remaining_branches": "Back up only owned settings.json, load the router page, corrupt that file, activate a saved own router, read status/toast, restore bytes immediately and doctor."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5914,+ "title": "[Bug]: Proxied Canvas app-backend responses carry two Date and two Server headers",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5914",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F20",+ "feature_ids": [+ "F20.app-backend-view"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "basis": "known-issue",+ "reason": "Canvas app-backend sessions forward an owned app backend HTTP response. Duplicate Date/Server response headers require an admitted bridge session and a reviewed backend emitting those headers; default demo without the bridge and a direct backend fetch do not prove the proxy contract. A Linux Agent Server is required; native macOS artifact backends are unsupported.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Canvas app-backend sessions forward an owned app backend HTTP response. Duplicate Date/Server response headers require an admitted bridge session and a reviewed backend emitting those headers; default demo without the bridge and a direct backend fetch do not prove the proxy contract. A Linux Agent Server is required; native macOS artifact backends are unsupported.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5910,+ "title": "[Bug]: Canvas backend prepare returns 500 'not a gzip file' for a non-gzip artifact instead of 409",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5910",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "source-consumer-review",+ "reason": "The reported malformed artifact prepare409-versus500 path has no current Canvas UI producer: CanvasExtensionsService offers install/enable/update/uninstall and bridge sessions but no backend/prepare operation. The backend admin REST contract belongs to the SDK; generic arbitrary app API access does not establish a mapped builtin consumer.",+ "current_result": "not-applicable",+ "fresh_reproduction_claim": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5899,+ "title": "[Bug]: Browser probe ignores PLAYWRIGHT_BROWSERS_PATH, so browser_tool_set is reported unusable and cannot start",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5899",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F13",+ "feature_ids": [+ "F13.openhands-options"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "current-source-awaiting-live",+ "reason": "Profile tool selection consumes the Agent Server tool catalog usability, including browser_tool_set. A genuine custom-only browser installation is required; standard user Chrome on this Mac would mask this condition.",+ "source_refs": [+ "src/constants/profile-tools.ts",+ "src/hooks/query/use-tool-catalog.ts",+ "SDKv1.55.0:openhands-tools/openhands/tools/browser_use/impl.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "A host/container with a genuine Chromium installed only under PLAYWRIGHT_BROWSERS_PATH, with no standard browser, default-cache browser, or PATH binary; do not remove the user Chrome install.",+ "map_note_status": "linked Known issue with exact prerequisite",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 131+ }+ ],+ "map_note_present": true,+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5897,+ "title": "[Bug]: Refreshing an installed skill or plugin whose source vanished, became invalid, or is 'local' returns 500 (expected 400/422, as install)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5897",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F19",+ "feature_ids": [+ "F19.refresh"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "live-reproduced",+ "reason": "Two real plugin Update clicks after moving only the owned source aside surfaced error toasts; refresh returned500 instead of handled400. Installed version/enabled state stayed unchanged. Source restored, own plugin uninstalled and fixture removed; no model call.",+ "source_refs": [+ "src/api/plugins-management-service.ts",+ "src/hooks/mutation/use-refresh-plugin.ts",+ "src/api/skills-service.ts"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "map_note_status": "current Known failure; executable block reproduced twice",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ [+ "control-openhands",+ "browser",+ "goto",+ "/plugins"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=plugins-add-plugin-button"+ ],+ [+ "control-openhands",+ "browser",+ "fill",+ "testid=add-plugin-source-input",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/workspace/qa-refresh-5897"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=add-plugin-submit"+ ],+ [+ "control-openhands",+ "browser",+ "wait",+ "testid=add-plugin-modal",+ "--state",+ "hidden",+ "--timeout",+ "30000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/plugins/installed"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=plugin-card-qa-refresh-5897"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=plugin-detail-refresh-qa-refresh-5897"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/plugins/installed/qa-refresh-5897/refresh",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--history"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/plugins/installed"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=plugin-detail-refresh-qa-refresh-5897"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/plugins/installed/qa-refresh-5897/refresh",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--history"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/plugins/installed"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/plugins/installed/qa-refresh-5897",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ]+ ],+ "observations": {+ "attempts": [+ {+ "attempt": 1,+ "status": 500,+ "version_unchanged": true,+ "enabled_unchanged": true,+ "ui_error_toast_present": true,+ "source_is_own_path": true,+ "source_is_local": false+ },+ {+ "attempt": 2,+ "status": 500,+ "version_unchanged": true,+ "enabled_unchanged": true,+ "ui_error_toast_present": true,+ "source_is_own_path": true,+ "source_is_local": false+ }+ ],+ "source_restored_before_uninstall": true+ },+ "artifacts": [+ "evidence/F19/5897-current.json"+ ],+ "no_successful_model_inference": true+ },+ "current_result": "fail",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 102+ }+ ],+ "map_note_present": true,+ "fresh_live_outcome": "mapped",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ [+ "control-openhands",+ "browser",+ "goto",+ "/plugins"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=plugins-add-plugin-button"+ ],+ [+ "control-openhands",+ "browser",+ "fill",+ "testid=add-plugin-source-input",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/workspace/qa-refresh-5897"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=add-plugin-submit"+ ],+ [+ "control-openhands",+ "browser",+ "wait",+ "testid=add-plugin-modal",+ "--state",+ "hidden",+ "--timeout",+ "30000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/plugins/installed"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=plugin-card-qa-refresh-5897"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=plugin-detail-refresh-qa-refresh-5897"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/plugins/installed/qa-refresh-5897/refresh",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--history"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/plugins/installed"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=plugin-detail-refresh-qa-refresh-5897"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/plugins/installed/qa-refresh-5897/refresh",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--history"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/plugins/installed"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/plugins/installed/qa-refresh-5897",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ]+ ],+ "artifacts": [+ "evidence/F19/5897-current.json"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced",+ "reproduction_setup_and_remaining_branches": "Install own local plugin, remove or invalidate only its source, click Update, record500 plus unchanged installed version/enabled state; restore source or uninstall fixture."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5894,+ "title": "[Bug]: POST /api/skills/install of a SKILL.md with unparsable YAML frontmatter returns 500 instead of 422",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5894",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas SkillsService only lists user/project skills plus bundled catalog. Add skill is an instructions-only /add-skill composer flow; Canvas does not call POST/api/skills/install. A malformed SDK management-install body is not a current direct Canvas consumer. The bundled add-skill skill instructs a GitHub fetch script writing workspace/.agents/skills; it does not use SDK management install.",+ "source_refs": [+ "src/api/skills-service.ts",+ "src/components/features/skills/add-skill-modal.tsx",+ "node_modules/@openhands/extensions/skills/add-skill/SKILL.md:16",+ "node_modules/@openhands/extensions/skills/add-skill/scripts/fetch_skill.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5890,+ "title": "[Bug]: managed LLM key refresh hook is lost after switch_llm / switch_profile, so the switched LLM's 401 is not refreshed",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5890",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F10",+ "feature_ids": [+ "F10.edit",+ "F10.default-used"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "current-source-awaiting-live",+ "reason": "Local conversation profile selection now fetches fresh profile config then calls switchLLM, so the managed refresh-hook loss after that switch remains a consumer. Refused-loopback errors cannot establish a real401 refresh/retry outcome.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts:1201",+ "src/hooks/mutation/use-switch-llm-profile.ts",+ "SDKv1.55.0:openhands-agent-server/openhands/agent_server/managed_llm_key.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "Approved managed-key refresh source and allow-listed real model endpoint that returns an actual401 and accepts the refreshed key; no fake successful model reply.",+ "map_note_status": "linked Known issue with exact prerequisite",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 178+ }+ ],+ "map_note_present": true,+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5889,+ "title": "[Bug]: A failed file upload overwrite destroys the previous file and leaves a partial one",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5889",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F05",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "feature_ids": [+ "F05.attach-files"+ ],+ "basis": "current-source-review",+ "reason": "Same-named Canvas attachment upload overwrites existing workspace file; a failed mid-stream upload can destroy previous contents.",+ "source_refs": [+ "src/api/conversation-file-upload.api.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Owned isolated backend with a controlled mid-upload filesystem failure (original Linuxprlimit file-size cap or equivalent bounded child-process limit), plus a pre-existing same-name file and a real Canvas overwrite. No such failure-injection backend has been arranged; do not change limits on shared runtime.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5883,+ "title": "[Bug]: GET /api/settings/secrets?agent_profile_id= answers 500 for an unloadable agent profile or a busy profile store (agent-profile routes answer 400/503)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5883",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas SecretsService calls SettingsClient.listSecrets() without agent_profile_id. Profile secret selection lists global names and saves secret_refs; the failing scoped secret-list query is not emitted.",+ "source_refs": [+ "src/api/secrets-service.ts:28",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5882,+ "title": "[Bug]: GET /api/llm/models?provider=deepseek|anthropic lists AWS Bedrock-only ids (deepseek.r1-v1:0, anthropic.claude-...-v1:0)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5882",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F10",+ "feature_ids": [+ "F10.model-picker",+ "F10.add-models"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "current-source-and-live-consumer-guard",+ "reason": "API retains Bedrock-only deepseek.* IDs, but current Canvas filters provider options to provider/ names: all3 affected IDs were absent from real Basic and Bulk add. SDK enumeration bug is not a current Canvas picker failure.",+ "source_refs": [+ "src/hooks/query/use-provider-models.ts",+ "src/components/shared/modals/settings/model-selector.tsx",+ "SDKv1.55.0:openhands-sdk/openhands/sdk/llm/utils/unverified_models.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "An own server with no AWS credentials; inspect GET/api/llm/models?provider=deepseek|anthropic and matching real Basic options, without saving or starting a model.",+ "map_note_status": "provisional note removed",+ "live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root",+ "artifacts": [+ "evidence/F10.model-picker/5882-consumer-correction.json"+ ],+ "observations": {+ "basic_bad_id_counts": [+ 0,+ 0,+ 0+ ],+ "bulk_bad_id_counts": {+ "deepseek.r1-v1:0": 0,+ "deepseek.v3-v1:0": 0,+ "deepseek.v3.2": 0+ }+ },+ "limitations": "SDK API bug still present; current Canvas only consumes prefixed rows"+ },+ "current_result": "pass",+ "fresh_live_outcome": "out-of-scope",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5881,+ "title": "[Bug]: POST /api/meta-profiles/{name} with neither classes nor prompt_template saves a router that can never route",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5881",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas MetaProfileEditor refuses Save unless classifier_model is nonempty and prompt_template contains{{instance_text}}; handleSave always sends classes:[] plus that template. It cannot produce the classless/template-less management body in this issue.",+ "source_refs": [+ "src/components/features/settings/meta-llm-profiles/meta-profile-editor.tsx:188"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5880,+ "title": "[Bug]: OpenAI subscription device/start and device/poll answer an unhandled 500 when auth.openai.com is unreachable or fails",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5880",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F10",+ "feature_ids": [+ "F10.subscription"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "live-reproduced",+ "reason": "Two real Connect clicks with the backend routed through an owned refused-loopback proxy returned500 rather than502/504, showing the friendly subscription error plus All connection attempts failed. UV_OFFLINE used the already-cached runtime. Each EXIT restored the original environment and restarted; doctor passed. No subscription account or model call; genuine pending-token poll remains untested.",+ "source_refs": [+ "src/api/llm-subscription-service.ts",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "SDKv1.55.0:openhands-agent-server/openhands/agent_server/llm_router.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "map_note_status": "current Known failure; executable block reproduced twice",+ "current_result": "fail",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ [+ "control-openhands",+ "bug",+ "run",+ "F10.subscription",+ "--name",+ "offline-start",+ "--attempts",+ "2",+ "--label",+ "current",+ "--timeout",+ "180",+ "--run",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root"+ ]+ ],+ "observations": {+ "device_start_statuses": [+ 500,+ 500+ ],+ "ui_failure_text": "Unable to connect your ChatGPT subscription. Try again or check the agent-server",+ "toast_text": "All connection attempts failed",+ "proxy": "owned refused loopback; local NO_PROXY",+ "cached_runtime_offline": true,+ "doctor_after_each": true,+ "pending_token_poll": "not tested",+ "setup_failure_excluded": "Initial proxy setup lacked UV_OFFLINE, so uvx attempted PyPI and restart failed before UI; no product verdict from it."+ },+ "artifacts": [+ "evidence/F10.subscription/5880-current.json",+ "evidence/settings-final-restoration.json"+ ],+ "no_successful_model_inference": true+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 119+ }+ ],+ "map_note_present": true,+ "fresh_live_outcome": "mapped",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ [+ "control-openhands",+ "bug",+ "run",+ "F10.subscription",+ "--name",+ "offline-start",+ "--attempts",+ "2",+ "--label",+ "current",+ "--timeout",+ "180",+ "--run",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root"+ ]+ ],+ "artifacts": [+ "evidence/F10.subscription/5880-current.json",+ "evidence/settings-final-restoration.json"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced",+ "reproduction_setup_and_remaining_branches": "Device-start exercised; poll still needs a genuine pending device token. No successful sign-in or inference is claimed."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5879,+ "title": "[Bug]: POST /api/meta-profiles/{name}/activate on ACP agent settings answers 200 \"activated\" while active_meta_profile stays null",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5879",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F12",+ "feature_ids": [+ "F12.activate"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F12-model-router.md",+ "basis": "live-reproduced",+ "reason": "With actual global ACP settings, two real Set active clicks each toasted activation success while API/settings active_meta_profile stayed null. Original owned settings restored byte-for-byte; no ACP subprocess or model call.",+ "source_refs": [+ "src/components/features/settings/meta-llm-profiles/meta-llm-settings-view.tsx:111",+ "src/hooks/mutation/use-activate-meta-profile.ts",+ "SDKv1.55.0:openhands-agent-server/openhands/agent_server/meta_profiles_router.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "map_note_status": "current Known failure; executable block reproduced twice",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ [+ "control-openhands",+ "api",+ "POST",+ "/api/meta-profiles/QA_ACP_ROUTER",+ "--data",+ "@/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/private/root-replays/router5879.json",+ "--write"+ ],+ [+ "control-openhands",+ "api",+ "PATCH",+ "/api/settings",+ "--data",+ "@/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/private/root-replays/globalacp5879.json",+ "--write",+ "--pick",+ "agent_settings.agent_kind"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/settings/meta-llm"+ ],+ [+ "control-openhands",+ "browser",+ "wait",+ "testid=meta-profile-row-QA_ACP_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-menu-trigger-QA_ACP_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-set-active"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Meta-profile \"QA_ACP_ROUTER\" activated",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/meta-profiles"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/settings"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-menu-trigger-QA_ACP_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-set-active"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Meta-profile \"QA_ACP_ROUTER\" activated",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/meta-profiles"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/settings"+ ],+ [+ "control-openhands",+ "browser",+ "screenshot",+ "--feature",+ "F12.activate",+ "--name",+ "oct16-acp-router"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/meta-profiles/QA_ACP_ROUTER",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ]+ ],+ "observations": {+ "attempts": [+ {+ "attempt": 1,+ "ui_success_toast": true,+ "active_meta_profile": null,+ "settings_active_meta_profile": null,+ "global_agent_kind": "acp"+ },+ {+ "attempt": 2,+ "ui_success_toast": true,+ "active_meta_profile": null,+ "settings_active_meta_profile": null,+ "global_agent_kind": "acp"+ }+ ],+ "settings_restored": true+ },+ "artifacts": [+ "evidence/F12/5879-current.json"+ ],+ "no_successful_model_inference": true+ },+ "current_result": "fail",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F12-model-router.md",+ "line": 85+ }+ ],+ "map_note_present": true,+ "fresh_live_outcome": "mapped",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ [+ "control-openhands",+ "api",+ "POST",+ "/api/meta-profiles/QA_ACP_ROUTER",+ "--data",+ "@/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/private/root-replays/router5879.json",+ "--write"+ ],+ [+ "control-openhands",+ "api",+ "PATCH",+ "/api/settings",+ "--data",+ "@/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/private/root-replays/globalacp5879.json",+ "--write",+ "--pick",+ "agent_settings.agent_kind"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/settings/meta-llm"+ ],+ [+ "control-openhands",+ "browser",+ "wait",+ "testid=meta-profile-row-QA_ACP_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-menu-trigger-QA_ACP_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-set-active"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Meta-profile \"QA_ACP_ROUTER\" activated",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/meta-profiles"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/settings"+ ],+ [+ "control-openhands",+ "browser",+ "toasts",+ "--clear"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-menu-trigger-QA_ACP_ROUTER"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=meta-profile-set-active"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Meta-profile \"QA_ACP_ROUTER\" activated",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/meta-profiles"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/settings"+ ],+ [+ "control-openhands",+ "browser",+ "screenshot",+ "--feature",+ "F12.activate",+ "--name",+ "oct16-acp-router"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/meta-profiles/QA_ACP_ROUTER",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ]+ ],+ "artifacts": [+ "evidence/F12/5879-current.json"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced",+ "reproduction_setup_and_remaining_branches": "Owned saved router plus backed-up actual global ACP settings; click Set active, compare toast and active_meta_profile readback, restore original global settings. No ACP subprocess/model turn needed."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5878,+ "title": "[Bug]: With OH_PERSISTENCE_DIR='~/x', meta-profiles are written to '<server cwd>/~/x/meta-profiles' instead of the expanded directory",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5878",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F12",+ "feature_ids": [+ "F12.create-first",+ "F12.list"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F12-model-router.md",+ "basis": "current-source-awaiting-live",+ "reason": "Router save/list uses the SDK MetaProfileStore, whose path helper mishandles a leading~ OH_PERSISTENCE_DIR. Default Canvas launcher uses an absolute owned path, so that normal run cannot exercise the report.",+ "source_refs": [+ "src/api/meta-profiles-service/meta-profiles-service.api.ts",+ "SDKv1.55.0:openhands-sdk/openhands/sdk/llm/meta_profile_store.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "A disposable Agent Server launched with literal leading-~ or relative OH_PERSISTENCE_DIR and private HOME/cwd; verify stored router path next to settings. Do not alter user persistence.",+ "map_note_status": "linked Known issue with exact prerequisite",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F12-model-router.md",+ "line": 134+ }+ ],+ "map_note_present": true,+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5873,+ "title": "[Bug]: Agent profile saved under a .json name mints a new id on every overwrite and leaves the active pointer dangling",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5873",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F13",+ "feature_ids": [+ "F13.create",+ "F13.edit",+ "F13.set-active"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "live-reproduced",+ "reason": "Real Canvas create/edit accepted QA_SUFFIX.json; two concurrency-only overwrites each minted a different UUID at revision0. On SDK1.55 the dangling active pointer repaired to default, so no persistent dangling-pointer claim. No model call.",+ "source_refs": [+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx:153",+ "src/utils/derive-profile-name.ts",+ "SDKv1.55.0:openhands-agent-server/openhands/agent_server/agent_profiles_router.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "map_note_status": "current Known failure; executable block reproduced twice",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/settings/agents"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=add-agent-profile"+ ],+ [+ "control-openhands",+ "browser",+ "fill",+ "testid=agent-profile-name-input",+ "QA_SUFFIX.json"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=save-agent-profile-btn"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Profile \"QA_SUFFIX.json\" created",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles/QA_SUFFIX"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-row >> has-text=QA_SUFFIX >> testid=agent-profile-menu-trigger"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-set-active"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Switched to profile \"QA_SUFFIX\"",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-row >> has-text=QA_SUFFIX >> testid=agent-profile-menu-trigger"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-edit"+ ],+ [+ "control-openhands",+ "browser",+ "fill",+ "testid=sdk-settings-tool_concurrency_limit",+ "2"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=save-agent-profile-btn"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Profile \"QA_SUFFIX.json\" updated",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles/QA_SUFFIX"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-row >> has-text=QA_SUFFIX >> testid=agent-profile-menu-trigger"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-edit"+ ],+ [+ "control-openhands",+ "browser",+ "fill",+ "testid=sdk-settings-tool_concurrency_limit",+ "3"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=save-agent-profile-btn"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Profile \"QA_SUFFIX.json\" updated",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles/QA_SUFFIX"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles"+ ],+ [+ "control-openhands",+ "browser",+ "screenshot",+ "--feature",+ "F13.edit",+ "--name",+ "oct16-json-profile"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/agent-profiles/a87f68c9-9109-432a-b0ad-3e255336ab15/activate",+ "--write"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/agent-profiles/QA_SUFFIX",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ]+ ],+ "observations": {+ "before": {+ "id": "db20bff3-fa85-49c0-b5ce-48f0aafc3215",+ "name": "QA_SUFFIX.json",+ "revision": 0+ },+ "overwrites": [+ {+ "id": "129ebe1b-7048-47a6-8628-0f3b05c92a5b",+ "name": "QA_SUFFIX.json",+ "revision": 0,+ "tool_concurrency_limit": 2,+ "active_agent_profile_id": "a87f68c9-9109-432a-b0ad-3e255336ab15"+ },+ {+ "id": "d8e7e7ea-964d-4c0b-986d-d7f05a65d977",+ "name": "QA_SUFFIX.json",+ "revision": 0,+ "tool_concurrency_limit": 3,+ "active_agent_profile_id": "a87f68c9-9109-432a-b0ad-3e255336ab15"+ }+ ]+ },+ "artifacts": [+ "evidence/F13/5873-current.json"+ ],+ "no_successful_model_inference": true+ },+ "current_result": "fail",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 89+ }+ ],+ "map_note_present": true,+ "fresh_live_outcome": "mapped",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/settings/agents"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=add-agent-profile"+ ],+ [+ "control-openhands",+ "browser",+ "fill",+ "testid=agent-profile-name-input",+ "QA_SUFFIX.json"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=save-agent-profile-btn"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Profile \"QA_SUFFIX.json\" created",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles/QA_SUFFIX"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-row >> has-text=QA_SUFFIX >> testid=agent-profile-menu-trigger"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-set-active"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Switched to profile \"QA_SUFFIX\"",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-row >> has-text=QA_SUFFIX >> testid=agent-profile-menu-trigger"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-edit"+ ],+ [+ "control-openhands",+ "browser",+ "fill",+ "testid=sdk-settings-tool_concurrency_limit",+ "2"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=save-agent-profile-btn"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Profile \"QA_SUFFIX.json\" updated",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles/QA_SUFFIX"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-row >> has-text=QA_SUFFIX >> testid=agent-profile-menu-trigger"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-edit"+ ],+ [+ "control-openhands",+ "browser",+ "fill",+ "testid=sdk-settings-tool_concurrency_limit",+ "3"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=save-agent-profile-btn"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Profile \"QA_SUFFIX.json\" updated",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles/QA_SUFFIX"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles"+ ],+ [+ "control-openhands",+ "browser",+ "screenshot",+ "--feature",+ "F13.edit",+ "--name",+ "oct16-json-profile"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/agent-profiles/a87f68c9-9109-432a-b0ad-3e255336ab15/activate",+ "--write"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/agent-profiles/QA_SUFFIX",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ]+ ],+ "artifacts": [+ "evidence/F13/5873-current.json"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced",+ "reproduction_setup_and_remaining_branches": "Create own QA_SUFFIX.json in Agent editor with valid LLM ref; activate, edit concurrency, Save and compare id/revision and active pointer. No model needed."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5872,+ "title": "[Bug]: TypeScript client's agent-profile payload types do not match what the server sends",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5872",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "This issue is the exported TypeScript type mirror: missing/extra declared fields and materialize diagnostics. Canvas does not call materialize or read resolved_mcp_servers; its editor preserves the real stored object through mergeAgentProfileSaveInput and sends disabled_skills-compatible fields. No concrete current rendered failure is supplied.",+ "source_refs": [+ "src/api/agent-profiles-service/agent-profiles-service.api.ts",+ "src/components/features/settings/agent-profiles/merge-agent-profile-save-input.ts"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5871,+ "title": "[Bug]: _seed_default_llm_profile docstring says materialize reports dangling_llm_profile_ref, but the materialize response has no such field",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5871",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "SDK internal _seed_default_llm_profile docstring promises a nonexistent materialize field. The report explicitly says runtime is fine; Canvas neither shows that docstring nor consumes materialize diagnostics.",+ "source_refs": [+ "src/api/agent-profiles-service/agent-profiles-service.api.ts"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5870,+ "title": "[Bug]: Bash event timestamp filters ignore the timezone offset (a +02:00 bound lands 2 h late)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5870",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "source-consumer-review",+ "reason": "Canvas BashService.searchEvents sends kind,command_id,sort_order,page_id,limit, never timestamp__gte/lt. Its event-history timestamp filters target conversation events, not BashEventService. The report's non-UTC Bash filter branch has no Canvas consumer.",+ "current_result": "not-applicable",+ "fresh_reproduction_claim": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5863,+ "title": "[Bug]: MCP OAuth callback URL with port 99999 or a non-numeric port answers 500 instead of 400",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5863",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas OAuth UI starts a probe, opens its authorization URL and polls status. It does not call the public submitOAuthCallback wrapper anywhere; malformed port URLs in that unused REST callback POST have no current UI entry point.",+ "source_refs": [+ "src/api/mcp-service/mcp-service.api.ts:317",+ "src/api/mcp-service/mcp-service.api.ts:292"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5862,+ "title": "[Bug]: Profile validate reports an invalid DeepSeek key (HTTP 401) as LLMBadRequestError instead of LLMAuthenticationError",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5862",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F10",+ "feature_ids": [+ "F10.create-validation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "current-source-awaiting-live",+ "reason": "Profile Save displays provider preflight error messages; its actual validation request uses this SDK exception mapping. A refused endpoint does not generate the real DeepSeek401 shape and cannot establish the error classification.",+ "source_refs": [+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx:380",+ "SDKv1.55.0:openhands-sdk/openhands/sdk/llm/exceptions/classifier.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "Explicit authorization for an actual invalid-key validation request to DeepSeek (no usable credentials/inference required); no such external request was authorized for this pass.",+ "map_note_status": "linked Known issue with exact prerequisite",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 179+ }+ ],+ "map_note_present": true,+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5861,+ "title": "[Bug]: DELETE /api/profiles/{name}.json deletes an LLM profile that an agent profile still cites (plain name answers 409)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5861",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas profile rows use names returned by the list and send those canonical names to Delete/Rename. The report bypasses FK guards by manually adding.json to the target URL; the UI has no alias-URL input and does not emit that alternate spelling.",+ "source_refs": [+ "src/api/profiles-service/profiles-service.api.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5860,+ "title": "[Bug]: POST /api/mcp/oauth/callback/{job_id} answers 500 'All connection attempts failed' once the job's callback server has closed",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5860",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "The double-submit/replay endpoint is not called by current Canvas OAuth UI. authorizeOAuth opens the provider popup and polls status; submitOAuthCallback has only its service definition/test callers. No current UI callback submission can be retried.",+ "source_refs": [+ "src/api/mcp-service/mcp-service.api.ts:317",+ "src/api/mcp-service/mcp-service.api.ts:429"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5849,+ "title": "[Bug]: workspaces write routes rewrite workspaces.json when nothing changes; a 404 DELETE creates the file and on read-only storage no-op writes answer 500",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5849",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F03",+ "feature_ids": [+ "F03.folder-browser",+ "F03.manage-workspaces"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "basis": "known-issue",+ "reason": "Canvas WorkspacesService add/remove workspace and parent calls back Home and new-conversation pickers. Read-only registry storage can turn a duplicate re-add or stale deletion into 500 instead of idempotent 200/404. File inode/mtime changes alone are not UI proof. Requires an owned read-only registry plus real duplicate/stale picker action; surface owns note, this ledger owns row.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Canvas WorkspacesService add/remove workspace and parent calls back Home and new-conversation pickers. Read-only registry storage can turn a duplicate re-add or stale deletion into 500 instead of idempotent 200/404. File inode/mtime changes alone are not UI proof. Requires an owned read-only registry plus real duplicate/stale picker action; surface owns note, this ledger owns row.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5847,+ "title": "[Bug]: PATCH /api/settings stores active_agent_profile_id / active_meta_profile pointers to profiles that do not exist",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5847",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas activates Agent/Model Router profiles through dedicated activate endpoints with known list IDs/names, rather than PATCHing arbitrary active_agent_profile_id/active_meta_profile pointers. The malformed unknown-pointer PATCH is not emitted by current settings saves.",+ "source_refs": [+ "src/hooks/mutation/use-activate-agent-profile.ts",+ "src/hooks/mutation/use-activate-meta-profile.ts",+ "src/api/settings-service/settings-service.api.ts"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5846,+ "title": "[Bug]: GET/DELETE /api/settings/secrets/{name} document 400 for an invalid name, but answer 422",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5846",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "OpenAPI docstrings advertise400 instead of422 for invalid secret names; runtime is already422. Canvas does not display these descriptions or branch on the documented400, and its form validates names. This is SDK documentation-only.",+ "source_refs": [+ "src/api/secrets-service.ts",+ "src/components/features/settings/secrets-settings/secret-form.tsx"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5845,+ "title": "[Bug]: VS Code URL's folder parameter is not URL-encoded, so a workspace path with '&' or '#' opens the wrong folder",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5845",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F08",+ "feature_ids": [+ "F08.vscode-link"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "Local VS Code controls call VSCodeClient.getUrl with the raw conversation workspace directory; reserved query characters can select the wrong folder. Requires a genuinely running editor exposed by this origin and an owned workspace containing &, #, ?, %, + or space. Cloud exposed_urls is a separate producer. Surface owns the note, this ledger owns the row.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Local VS Code controls call VSCodeClient.getUrl with the raw conversation workspace directory; reserved query characters can select the wrong folder. Requires a genuinely running editor exposed by this origin and an owned workspace containing &, #, ?, %, + or space. Cloud exposed_urls is a separate producer. Surface owns the note, this ledger owns the row.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5844,+ "title": "[Bug]: in Docker runtime mode, the slashless workspace root rejects the workspace cookie with 401",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5844",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "source-consumer-review",+ "reason": "The reported401 is specific to the slashless Docker workspace root. Canvas useWorkspaceSession receives a trailing-slash base from RemoteWorkspace.startWorkspaceSession; joinWorkspaceUrl preserves it for empty and file paths. It never generates the affected slashless root.",+ "current_result": "not-applicable",+ "fresh_reproduction_claim": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5843,+ "title": "[Bug]: GET /api/git/changes for a repository subdirectory mixes repository-relative and subdirectory-relative paths",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5843",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F08",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "feature_ids": [+ "F08.uncommitted-changes",+ "F08.diff-view-modes"+ ],+ "basis": "live-reproduced",+ "reason": "Canvas Commits in repo/src showed mixed paths README.md,new.py,src/app.py,src/util.py. Expanding src/app.py requested repo/src/src/app.py and returned400 twice, so the real file diff could not load. Root repo control showed five correctly based paths.",+ "source_refs": [+ "src/hooks/query/use-unified-get-git-changes.ts",+ "src/hooks/query/use-unified-git-diff.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": {+ "number": 5843,+ "result": "fail",+ "actual": "Canvas Commits in repo/src showed mixed paths README.md,new.py,src/app.py,src/util.py. Expanding src/app.py requested repo/src/src/app.py and returned400 twice, so the real file diff could not load. Root repo control showed five correctly based paths.",+ "expected": "Subdirectory workspace lists only files beneath src relative to src and can open their real diff.",+ "evidence": [+ "evidence/F08.diff-view-modes/subdir-mixed-paths-5843.png"+ ],+ "steps": [+ "Owned repo commits README.md,src/app.py,src/util.py; modify README/app, add root notes.txt and src/new.py, remove util.py.",+ "Create idle dummy conversation with working_dir=<repo>/src; open right panel \u2192 Commits \u2192 Uncommitted.",+ "Expand displayed src/app.py; actual /git/diff path is <repo>/src/src/app.py and status400."+ ],+ "at": "2026-10-16T04:43:26.493899+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F08.uncommitted-changes",+ "F08.diff-view-modes"+ ],+ "known_bug_contract": {+ "id": "F08.diff-view-modes",+ "name": "subdirectory-paths",+ "status": "xfail",+ "sha": "2ab83809aa659dcc",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "400",+ "bugShows": "400",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"400\"",+ "actual": "400",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F08.diff-view-modes-subdirectory-paths-1792126467938-1.json"+ }+ },+ "live_recipe": "Own repo: root README modified, src/app.py modified, src/new.py untracked. Select repo/src workspace; open Commits, compare root/subdir changes and open listed diff.",+ "known_bug_contract": {+ "id": "F08.diff-view-modes",+ "name": "subdirectory-paths",+ "status": "xfail",+ "sha": "2ab83809aa659dcc",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "400",+ "bugShows": "400",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"400\"",+ "actual": "400",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F08.diff-view-modes-subdirectory-paths-1792126467938-1.json"+ },+ "current_result": "fail",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Owned repo commits README.md,src/app.py,src/util.py; modify README/app, add root notes.txt and src/new.py, remove util.py.",+ "Create idle dummy conversation with working_dir=<repo>/src; open right panel \u2192 Commits \u2192 Uncommitted.",+ "Expand displayed src/app.py; actual /git/diff path is <repo>/src/src/app.py and status400."+ ],+ "artifacts": [+ "evidence/F08.diff-view-modes/subdir-mixed-paths-5843.png"+ ],+ "observation": "Canvas Commits in repo/src showed mixed paths README.md,new.py,src/app.py,src/util.py. Expanding src/app.py requested repo/src/src/app.py and returned400 twice, so the real file diff could not load. Root repo control showed five correctly based paths.",+ "limit": null,+ "no_successful_model_inference": true+ },+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Owned repo commits README.md,src/app.py,src/util.py; modify README/app, add root notes.txt and src/new.py, remove util.py.",+ "Create idle dummy conversation with working_dir=<repo>/src; open right panel \u2192 Commits \u2192 Uncommitted.",+ "Expand displayed src/app.py; actual /git/diff path is <repo>/src/src/app.py and status400."+ ],+ "artifacts": [+ "evidence/F08.diff-view-modes/subdir-mixed-paths-5843.png"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5841,+ "title": "[Bug]: git diff of a file that git changes lists as DELETED returns 400 'File does not exist'",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5841",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F08",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "feature_ids": [+ "F08.diff-deleted-file"+ ],+ "basis": "scoped-live-not-reproduced",+ "reason": "Actual Canvas Commits lists deleted src/util.py; expanding it displays the defined deleted-file placeholder and sends zero /git/diff requests. Source intentionally disables working-tree D diff. SDK endpoint400 is bypassed by this consumer; no claim server fixed.",+ "source_refs": [+ "src/hooks/query/use-unified-git-diff.ts",+ "src/components/features/conversation/conversation-tabs/"+ ],+ "issue_state": "open",+ "fresh_live_outcome": {+ "number": 5841,+ "result": "pass",+ "outcome": "out-of-scope",+ "actual": "Actual Canvas Commits lists deleted src/util.py; expanding it displays the defined deleted-file placeholder and sends zero /git/diff requests. Source intentionally disables working-tree D diff. SDK endpoint400 is bypassed by this consumer; no claim server fixed.",+ "expected": "Current F08.diff-deleted-file renders its deleted-file fallback without a failing diff request.",+ "evidence": [+ "evidence/F08.diff-deleted-file/deleted-fallback-5841.png",+ "evidence/F08.diff-deleted-file/deleted-fallback-5841.aria.txt"+ ],+ "steps": [+ "Create disposable Git repo with committed src/util.py, remove only that file, create own idle conversation in repo.",+ "Open right panel \u2192 Commits \u2192 Uncommitted; clear network.",+ "Expand src/util.py via file-diff-viewer-outer containing that path \u2192 collapse button.",+ "Wait exact deleted-file placeholder. browser network --filter /git/diff returns total0."+ ],+ "limit": "No backend-only status proof; this specific SDK route contract has no active working-tree deletion request in Canvas.",+ "at": "2026-10-16T04:42:28.722021+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F08.diff-deleted-file"+ ]+ },+ "live_recipe": "Commit owned src/util.py then delete it; open deleted row in Commits, capture UI and /git/diff status; committed file is control.",+ "current_result": "pass",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Create disposable Git repo with committed src/util.py, remove only that file, create own idle conversation in repo.",+ "Open right panel \u2192 Commits \u2192 Uncommitted; clear network.",+ "Expand src/util.py via file-diff-viewer-outer containing that path \u2192 collapse button.",+ "Wait exact deleted-file placeholder. browser network --filter /git/diff returns total0."+ ],+ "artifacts": [+ "evidence/F08.diff-deleted-file/deleted-fallback-5841.png",+ "evidence/F08.diff-deleted-file/deleted-fallback-5841.aria.txt"+ ],+ "observation": "Actual Canvas Commits lists deleted src/util.py; expanding it displays the defined deleted-file placeholder and sends zero /git/diff requests. Source intentionally disables working-tree D diff. SDK endpoint400 is bypassed by this consumer; no claim server fixed.",+ "limit": "No backend-only status proof; this specific SDK route contract has no active working-tree deletion request in Canvas.",+ "no_successful_model_inference": true+ },+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Create disposable Git repo with committed src/util.py, remove only that file, create own idle conversation in repo.",+ "Open right panel \u2192 Commits \u2192 Uncommitted; clear network.",+ "Expand src/util.py via file-diff-viewer-outer containing that path \u2192 collapse button.",+ "Wait exact deleted-file placeholder. browser network --filter /git/diff returns total0."+ ],+ "artifacts": [+ "evidence/F08.diff-deleted-file/deleted-fallback-5841.png",+ "evidence/F08.diff-deleted-file/deleted-fallback-5841.aria.txt"+ ],+ "needs_current_replay": false,+ "current_verdict": "scope-reviewed",+ "fresh_reproduction_claim": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5839,+ "title": "[Bug]: switch_profile reinstalls the copy cached at the first switch after the profile file was edited",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5839",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Current Local Canvas already fetches fresh config and invokes switchLLM with a fresh profile:<name>:<uuidv4()> usage_id, deliberately avoiding the SDK switch_profile registry cache. Cloud uses its app-server route, a different transport from this Agent Server reproducer.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts:1201"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5838,+ "title": "[Bug]: after a mid-run switch_llm answers 200, GET and base_state.json keep the old LLM until the step ends",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5838",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F10",+ "feature_ids": [+ "F10.default-used"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "current-source-awaiting-live",+ "reason": "Canvas live profile selection calls switchLLM on the running conversation and current source does not force persisted readback. The GET/base_state disagreement during a real running tool step remains relevant.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts:1201",+ "src/hooks/mutation/use-switch-llm-profile.ts",+ "SDKv1.55.0:openhands-sdk/openhands/sdk/conversation/impl/local_conversation.py"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "prerequisite": "A genuinely running model-issued bounded tool step (for example sleep20) and approved starting model; refused-loopback initialization never enters that state.",+ "map_note_status": "linked Known issue with exact prerequisite",+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 180+ }+ ],+ "map_note_present": true,+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5837,+ "title": "[Bug]: GET /api/file/search_subdirs pagination never gets past two directories whose names differ only by case",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5837",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F03",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "feature_ids": [+ "F03.folder-browser"+ ],+ "basis": "current-source-review",+ "reason": "Workspace folder picker follows search_subdirs cursor; case-only directory twins can repeat cursor and hide later choices.",+ "source_refs": [+ "src/hooks/query/use-search-subdirs.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Case-sensitive filesystem with distinct case-only names at actual picker pagination boundary; default case-insensitive Mac volume cannot hold fixture.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5836,+ "title": "[Bug]: File upload responds only {\"success\": true}, so the SDK and TypeScript remote workspaces report no file_size",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5836",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Canvas awaits upload success but ignores file_size, building uploaded/skipped names from File inputs. TS LocalConversation write_file is not Canvas agent runtime.",+ "source_refs": [+ "src/api/conversation-file-upload.api.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5835,+ "title": "[Bug]: idle eviction kills a running conversation-scoped bash command and writes no final BashOutput",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5835",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F27",+ "feature_ids": [+ "F27.overview-identity"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "known-issue",+ "reason": "Local Overview Git reads use a conversation-scoped BashClient.executeCommand with a10s timeout. With a short configured idle TTL and an overlapping eviction pass, scoped work can be killed without final output. Recheck a genuine still-running Canvas Git probe on that owned runtime; agent Terminal events are a different producer from the standalone async sleep in the report.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Local Overview Git reads use a conversation-scoped BashClient.executeCommand with a10s timeout. With a short configured idle TTL and an overlapping eviction pass, scoped work can be killed without final output. Recheck a genuine still-running Canvas Git probe on that owned runtime; agent Terminal events are a different producer from the standalone async sleep in the report.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5834,+ "title": "[Bug]: GET /api/file/archive `format` description promises a tar.gz of \"the entire directory\", but a directory holding one repository archives only that repository",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5834",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Explicitly an OpenAPI/generated description bug; Canvas does not display that endpoint prose. Runtime behavior is intentionally unchanged.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5833,+ "title": "[Bug]: Conversation-scoped bash routes return 500 \"'State' object has no attribute 'runtime_event_service'\" for a non-UUID conversation id (expected 422)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5833",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "source-consumer-review",+ "reason": "The500 branch requires a malformed non-UUID conversation id. Canvas gets its scoped Bash id from a real active conversation and route guards resolve invalid routes before useLocalGitInfo enables; the builtin producer does not issue this malformed-id call. SDK parameter validation is independent.",+ "current_result": "not-applicable",+ "fresh_reproduction_claim": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5825,+ "title": "[Bug]: A credential binding sent while a regular (non-ACP) conversation is unloaded makes it require CODEX_AUTH_JSON re-activation after every pause",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5825",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas does not issue PUT credential-bindings/CODEX_AUTH_JSON to regular unloaded conversations. Native secret delivery uses structured conversation secrets; ACP binding lifecycle is server/Cloud infrastructure and the report depends on an externally injected non-ACP binding.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/agent-server-adapter.ts"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5824,+ "title": "[Bug]: A goal resumed after a failed judge audit ends \"capped\" at iteration 2 of max_iterations 1",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5824",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F05",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "feature_ids": [+ "F05.slash-goal"+ ],+ "basis": "current-source-review",+ "reason": "Canvas exposes /goal --max and iteration/max banner; failed judge then Resume can show2/1 and overrun cap.",+ "source_refs": [+ "src/hooks/chat/use-goal-interceptor.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Working goal agent/judge, controlled judge failure, then successful incomplete verdict; no provider inference authorized.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5823,+ "title": "[Bug]: live switch_acp_model to a model the ACP provider does not offer answers 500 instead of a 4xx",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5823",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Current ACP composer picker lists only availableAcpModels from the live session/discovery and has no free-text unknown-model switch input. The reported request manually sends an unoffered ID to the API.",+ "source_refs": [+ "src/hooks/use-chat-input-model-state.ts:122",+ "src/components/features/chat/components/chat-input-actions.tsx",+ "src/components/features/chat/components/chat-input-model.tsx:84"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5822,+ "title": "[Bug]: agent_final_response still returns the abandoned branch's answer after navigate moves HEAD back",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5822",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Canvas never calls agent_final_response nor server navigate. Message editing forks instead, so reported abandoned-branch route has no consumer.",+ "source_refs": [+ "src/hooks/mutation/use-fork-conversation.ts",+ "src/api/conversation-service/agent-server-conversation-service.api.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5820,+ "title": "[Bug]: POST /condense on an agent with no request-handling condenser returns 500 with a traceback instead of a 4xx",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5820",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F05",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "basis": "live-blocked",+ "reason": "Compact context is exposed without checking server condenser support; null/no-op condenser can return500.",+ "source_refs": [+ "src/hooks/use-compact-context-action.ts",+ "src/hooks/mutation/use-condense-conversation.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5820,+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "result": "blocked",+ "at": "2026-10-16T04:38:33.712264+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "actual": "Actual idle Usage panel has cost0 but usage=null, so no context card/Compact context action. A known gpt-4o dummy model with max_input_tokens128000 and genuine refused-loopback send still yields no context control. Separate direct API diagnostic returned500 No condenser configured; this is not claimed as UI reproduction.",+ "expected": "A reachable Compact context action reports unsupported condenser as a client error.",+ "prerequisite": "Genuine populated token-usage/context metrics on a null/no-op-condenser conversation, requiring successful provider response or an existing recorded real conversation; do not inject fabricated stats.",+ "evidence": [+ "evidence/F05.context-window-meter/condenser-ui-prerequisite-5820.png"+ ],+ "steps": [+ "Create idle tools-empty offline conversation with condenser:null.",+ "Open right panel; choose Usage; inspect absence of Compact context.",+ "Control: known model openai/gpt-4o/max_input_tokens128000, base_url http://127.0.0.1:1/v1, dummy key; send once. Status error and still no context controls.",+ "Separate backend diagnostic only: POST /api/conversations/<idle-id>/condense returned500 No condenser configured."+ ]+ },+ "live_recipe": "Create idle offline conversation with condenser:null; click context-window-meter then context-window-compact-button; inspect genuine POST/condense and toast, no model.",+ "verification": "blocked",+ "prerequisite": "Genuine populated token-usage/context metrics on a null/no-op-condenser conversation, requiring successful provider response or an existing recorded real conversation; do not inject fabricated stats.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5813,+ "title": "[Bug]: POST /api/conversations/{id}/ask_agent returns 500 for an unknown conversation instead of the documented 404",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5813",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F05",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "feature_ids": [+ "F05.slash-btw"+ ],+ "basis": "scoped-live-not-reproduced",+ "reason": "Actual /btw submission on an owned conversation deleted out of band made a200 batch metadata preflight returning missing conversation, then zero ask_agent requests. fetchConversationData throws V1 conversation not found before the affected route. After metadata polling Canvas returned to list. No assertion that SDK unknown-ID endpoint is fixed.",+ "source_refs": [+ "src/hooks/chat/use-btw-interceptor.ts",+ "src/hooks/mutation/conversation-mutation-utils.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": {+ "number": 5813,+ "result": "pass",+ "outcome": "out-of-scope",+ "actual": "Actual /btw submission on an owned conversation deleted out of band made a200 batch metadata preflight returning missing conversation, then zero ask_agent requests. fetchConversationData throws V1 conversation not found before the affected route. After metadata polling Canvas returned to list. No assertion that SDK unknown-ID endpoint is fixed.",+ "expected": "Stale /btw must not invoke an unknown conversation as a valid agent.",+ "evidence": [+ "evidence/F05.slash-btw/deleted-conversation-5813.png"+ ],+ "steps": [+ "Open own idle conversation and fill /btw QA_SURFACE stale check.",+ "Delete exactly that owned ID through native API, then press Enter in chat-input.",+ "Inspect scoped network: GET /api/conversations?ids=<deleted>200; ask_agent total0. Source fetchConversationData rejects null preflight before calling askAgent."+ ],+ "limit": "Does not cover deletion in the micro-window between successful preflight and ask_agent; no such controlled concurrent transport fixture was arranged.",+ "at": "2026-10-16T04:46:23.398005+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F05.slash-btw"+ ]+ },+ "live_recipe": "Open own idle conversation, delete via API without refetch/navigation, submit /btw hi. Inspect ask_agent status/toast; no model reached.",+ "current_result": "pass",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Open own idle conversation and fill /btw QA_SURFACE stale check.",+ "Delete exactly that owned ID through native API, then press Enter in chat-input.",+ "Inspect scoped network: GET /api/conversations?ids=<deleted>200; ask_agent total0. Source fetchConversationData rejects null preflight before calling askAgent."+ ],+ "artifacts": [+ "evidence/F05.slash-btw/deleted-conversation-5813.png"+ ],+ "observation": "Actual /btw submission on an owned conversation deleted out of band made a200 batch metadata preflight returning missing conversation, then zero ask_agent requests. fetchConversationData throws V1 conversation not found before the affected route. After metadata polling Canvas returned to list. No assertion that SDK unknown-ID endpoint is fixed.",+ "limit": "Does not cover deletion in the micro-window between successful preflight and ask_agent; no such controlled concurrent transport fixture was arranged.",+ "no_successful_model_inference": true+ },+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Open own idle conversation and fill /btw QA_SURFACE stale check.",+ "Delete exactly that owned ID through native API, then press Enter in chat-input.",+ "Inspect scoped network: GET /api/conversations?ids=<deleted>200; ask_agent total0. Source fetchConversationData rejects null preflight before calling askAgent."+ ],+ "artifacts": [+ "evidence/F05.slash-btw/deleted-conversation-5813.png"+ ],+ "needs_current_replay": false,+ "current_verdict": "scope-reviewed",+ "fresh_reproduction_claim": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5810,+ "title": "[Bug]: Loading the same plugin twice duplicates its hooks, so they run twice per message",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5810",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "No current Canvas caller emits /load_plugin. Plugins are attached during conversation creation and read in metadata; the issue requires calling the dynamic load_plugin API twice. Creation-time plugin merging is a different path and not established by this report.",+ "source_refs": [+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/use-conversation-plugins.ts"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5809,+ "title": "[Bug]: Session socket delivers a user message after the last_user_message_id update it causes (seq 0, 2, 1)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5809",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Canvas already tracks highest contiguous sequence, not highest seen. For0,2,1 its cursor stays0 until1 arrives, preventing reported reconnect skip; no claim wire ordering fixed.",+ "source_refs": [+ "src/utils/session-seq-cursor.ts",+ "src/contexts/conversation-websocket-context.tsx"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5808,+ "title": "[Bug]: TS client switchAcpModel docstrings say the server returns 409 before the first message, but it answers 200 and defers the switch",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5808",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "SDK TypeScript docstrings incorrectly promise409 pre-first-run; Canvas adapter already documents200 deferred behavior and does not rely on those stale comments. Runtime is explicitly unchanged.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts:1222"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5807,+ "title": "[Bug]: switch_acp_model before the first run accepts a blank or whitespace model and persists it",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5807",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "ACP composer switches only from concrete model entries returned by its session/discovery (or the explicit Agent default path); it has no blank/free-text model switch input. The reported whitespace-only REST body is not emitted by this current consumer; settings profile creation is a different API path.",+ "source_refs": [+ "src/hooks/use-chat-input-model-state.ts",+ "src/hooks/mutation/use-switch-acp-model.ts",+ "src/constants/acp-providers.ts",+ "src/components/features/chat/components/chat-input-model.tsx:84"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5806,+ "title": "[Bug]: GET /api/conversations/{id}/events/{event_id} with an unknown id returns 500 instead of 404, and batch get fails whole on one missing id",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5806",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.branch-from-here"+ ],+ "basis": "current-source-review",+ "reason": "User-message edit/fork reads parent with getEvent; displayed event whose file disappears can fail500 before forking. Canvas does not use batch event getter.",+ "source_refs": [+ "src/hooks/mutation/use-fork-conversation.ts",+ "src/api/conversation-service/agent-server-conversation-service.api.ts:getEventParentId"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Owned persisted conversation and displayed user event whose event file becomes missing, then actual edit/fork; random API unknown-ID check alone is not Canvas proof.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5804,+ "title": "[Bug]: POST /api/conversations/{id}/events with a non-user role returns 500 AssertionError instead of a 4xx",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5804",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Supported Canvas send paths always send user role; no UI posts assistant/system/tool role through SendMessageRequest.",+ "source_refs": [+ "src/hooks/chat/use-chat-submission.ts",+ "src/api/conversation-service/agent-server-conversation-service.api.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5801,+ "title": "[Bug]: Event log is not in timestamp order: stats updates land after their step with an earlier timestamp, so timestamp__gte and resend_mode=since skip them",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5801",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Reported loss requires timestamp__gte continuation or legacy resend_mode=since. Current reconnect uses session after_seq and unfiltered count; timestampLt older paging is a distinct contract.",+ "source_refs": [+ "src/utils/session-seq-cursor.ts",+ "src/contexts/conversation-websocket-context.tsx",+ "src/hooks/use-load-older-events.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5800,+ "title": "[Bug]: Unknown conversation ids get 400 on /pause, /interrupt and DELETE and 200 {\"success\": false} on PATCH instead of the documented 404",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5800",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F07",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "feature_ids": [+ "F07.delete-confirm",+ "F07.rename-inline"+ ],+ "basis": "live-reproduced",+ "reason": "Actual Delete confirmation after out-of-band removal issuedDELETE400 and showed Bad Request. Stale inline rename control PATCH200 was followed by missing metadata and explicit not-found UI, so Canvas does not falsely confirm that rename.",+ "source_refs": [+ "src/api/conversation-service/agent-server-conversation-service.api.ts"+ ],+ "issue_state": "open",+ "fresh_live_outcome": {+ "number": 5800,+ "result": "fail",+ "actual": "Actual Delete confirmation after out-of-band removal issuedDELETE400 and showed Bad Request. Stale inline rename control PATCH200 was followed by missing metadata and explicit not-found UI, so Canvas does not falsely confirm that rename.",+ "expected": "An already-deleted conversation is reported as missing rather than generic Bad Request; stale rename must not claim success.",+ "evidence": [+ "evidence/F07.delete-confirm/stale-delete-5800.png",+ "evidence/F07.rename-inline/stale-rename-5800.png"+ ],+ "steps": [+ "Open own idle conversation \u2192 header More options \u2192 Delete to leave confirmation open.",+ "Delete same owned ID through native API; click actual Confirm Delete. Network DELETE400; toast Bad Request.",+ "Separate own idle conversation: begin inline rename, delete same ID via API, press Enter. PATCH200, then GET batch returns missing and UI shows not-found without success toast."+ ],+ "limit": "Pause/interrupt variants not driven; only Delete reproduces the generic400 symptom, and rename has a current UI safeguard.",+ "at": "2026-10-16T04:48:05.989823+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F07.delete-confirm",+ "F07.rename-inline"+ ],+ "known_bug_contract": {+ "id": "F07.delete-confirm",+ "name": "already-deleted",+ "status": "xfail",+ "sha": "8c72c361f9674a68",+ "assertion": {+ "kind": "mismatch",+ "expected": "404",+ "actual": "400",+ "bugShows": "400",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"400\"",+ "actual": "400",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F07.delete-confirm-already-deleted-1792126504524-1.json"+ }+ },+ "live_recipe": "Keep own idle row/page open, delete same ID via own API without UI refetch, use visible Delete confirmation and inspect400/toast. Rename PATCH200 is followed by404 GET.",+ "known_bug_contract": {+ "id": "F07.delete-confirm",+ "name": "already-deleted",+ "status": "xfail",+ "sha": "8c72c361f9674a68",+ "assertion": {+ "kind": "mismatch",+ "expected": "404",+ "actual": "400",+ "bugShows": "400",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"400\"",+ "actual": "400",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F07.delete-confirm-already-deleted-1792126504524-1.json"+ },+ "current_result": "fail",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Open own idle conversation \u2192 header More options \u2192 Delete to leave confirmation open.",+ "Delete same owned ID through native API; click actual Confirm Delete. Network DELETE400; toast Bad Request.",+ "Separate own idle conversation: begin inline rename, delete same ID via API, press Enter. PATCH200, then GET batch returns missing and UI shows not-found without success toast."+ ],+ "artifacts": [+ "evidence/F07.delete-confirm/stale-delete-5800.png",+ "evidence/F07.rename-inline/stale-rename-5800.png"+ ],+ "observation": "Actual Delete confirmation after out-of-band removal issuedDELETE400 and showed Bad Request. Stale inline rename control PATCH200 was followed by missing metadata and explicit not-found UI, so Canvas does not falsely confirm that rename.",+ "limit": "Pause/interrupt variants not driven; only Delete reproduces the generic400 symptom, and rename has a current UI safeguard.",+ "no_successful_model_inference": true+ },+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Open own idle conversation \u2192 header More options \u2192 Delete to leave confirmation open.",+ "Delete same owned ID through native API; click actual Confirm Delete. Network DELETE400; toast Bad Request.",+ "Separate own idle conversation: begin inline rename, delete same ID via API, press Enter. PATCH200, then GET batch returns missing and UI shows not-found without success toast."+ ],+ "artifacts": [+ "evidence/F07.delete-confirm/stale-delete-5800.png",+ "evidence/F07.rename-inline/stale-rename-5800.png"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5799,+ "title": "[Bug]: PATCH /api/conversations/{id} with a whitespace-only title returns 200 and erases the stored title",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5799",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "current-source-review",+ "reason": "Canvas trims title before PATCH, so whitespace UI input sends empty string and takes422, rather than SDK whitespace200. Existing Canvas17061 owns blank-rename UI behavior.",+ "source_refs": [+ "src/components/features/conversation/conversation-name.tsx"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5791,+ "title": "[Bug]: A secret value that is not an object returns 500 \"'str'/'int' object has no attribute 'pop'\"",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5791",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas constructs structured secret source objects for conversation creation and sends string values only to the update endpoint that intentionally accepts them. Number/null/boolean malformed values and plain-string create bodies are not current UI payloads.",+ "source_refs": [+ "src/api/agent-server-adapter.ts",+ "src/api/conversation-service/agent-server-conversation-service.api.ts",+ "src/api/secrets-service.ts"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5790,+ "title": "[Bug]: GET /api/conversations with 100 or more ids returns 500 with an empty exception instead of a 4xx naming the limit",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5790",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "basis": "live-reproduced",+ "reason": "Real99-child parent load issued200 batch and opened planner plus parent session sockets. Adding the100th genuine idle child then full reload issued500 batch, displayed Internal Server Error and opened only parent socket, losing discovered planner. No model runs; no browser storage read.",+ "source_refs": [+ "src/hooks/query/use-sub-conversations.ts",+ "src/contexts/websocket-provider-wrapper.tsx",+ "src/hooks/use-handle-plan-click.ts",+ "src/api/conversation-service/agent-server-conversation-service.api.ts:batchGetAppConversations"+ ],+ "issue_state": "open",+ "fresh_live_outcome": {+ "number": 5790,+ "result": "fail",+ "actual": "Real99-child parent load issued200 batch and opened planner plus parent session sockets. Adding the100th genuine idle child then full reload issued500 batch, displayed Internal Server Error and opened only parent socket, losing discovered planner. No model runs; no browser storage read.",+ "expected": "Planner discovery remains available with100 genuine child IDs; requests obey server limit or fail as an explicit client error.",+ "evidence": [+ "evidence/F06.tool-visualizers/planner-99-control-5790.png",+ "evidence/F06.tool-visualizers/planner-100-batch-5790.png",+ "evidence/F06.tool-visualizers/planner-5790-summary.json"+ ],+ "steps": [+ "Create own tools-empty offline parent and99 idle children with same workspace and real parent_conversation_id; one child tags.plannerparent=<parent>.",+ "In newly initialized configured browser, goto parent; network ids= batch200; lifecycle metadata opens planner and parent.",+ "Create100th native idle child. Full reload resets in-memory planner discovery; no UI-created planner hint was ever arranged.",+ "Actual network batch500; UI toast Internal Server Error; lifecycle metadata shows only new parent connection."+ ],+ "at": "2026-10-16T04:40:32.069176+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "known_bug_contract": {+ "id": "F06.tool-visualizers",+ "name": "planner-batch-100",+ "status": "xfail",+ "sha": "e9b66a642e0d7f95",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "500",+ "bugShows": "500",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"500\"",+ "actual": "500",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F06.tool-visualizers-planner-batch-100-1792126521614-1.json"+ }+ },+ "live_recipe": "Create own idle parent and100 genuine idle children via parent_conversation_id (no runs), including one child with real plannerparent tag. Reload without own trusted planning ID and compare99/100-ID batch; assert planner discovery/stream rather than nonexistent generic child-list component.",+ "known_bug_contract": {+ "id": "F06.tool-visualizers",+ "name": "planner-batch-100",+ "status": "xfail",+ "sha": "e9b66a642e0d7f95",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "500",+ "bugShows": "500",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"500\"",+ "actual": "500",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F06.tool-visualizers-planner-batch-100-1792126521614-1.json"+ },+ "current_result": "fail",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Create own tools-empty offline parent and99 idle children with same workspace and real parent_conversation_id; one child tags.plannerparent=<parent>.",+ "In newly initialized configured browser, goto parent; network ids= batch200; lifecycle metadata opens planner and parent.",+ "Create100th native idle child. Full reload resets in-memory planner discovery; no UI-created planner hint was ever arranged.",+ "Actual network batch500; UI toast Internal Server Error; lifecycle metadata shows only new parent connection."+ ],+ "artifacts": [+ "evidence/F06.tool-visualizers/planner-99-control-5790.png",+ "evidence/F06.tool-visualizers/planner-100-batch-5790.png",+ "evidence/F06.tool-visualizers/planner-5790-summary.json"+ ],+ "observation": "Real99-child parent load issued200 batch and opened planner plus parent session sockets. Adding the100th genuine idle child then full reload issued500 batch, displayed Internal Server Error and opened only parent socket, losing discovered planner. No model runs; no browser storage read.",+ "limit": null,+ "no_successful_model_inference": true+ },+ "run_id": "backfill-18334-refresh",+ "steps": [+ "Create own tools-empty offline parent and99 idle children with same workspace and real parent_conversation_id; one child tags.plannerparent=<parent>.",+ "In newly initialized configured browser, goto parent; network ids= batch200; lifecycle metadata opens planner and parent.",+ "Create100th native idle child. Full reload resets in-memory planner discovery; no UI-created planner hint was ever arranged.",+ "Actual network batch500; UI toast Internal Server Error; lifecycle metadata shows only new parent connection."+ ],+ "artifacts": [+ "evidence/F06.tool-visualizers/planner-99-control-5790.png",+ "evidence/F06.tool-visualizers/planner-100-batch-5790.png",+ "evidence/F06.tool-visualizers/planner-5790-summary.json"+ ],+ "needs_current_replay": false,+ "current_verdict": "reproduced",+ "fresh_reproduction_claim": true,+ "verification_status": "reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5789,+ "title": "[Bug]: allow_cors_origins delivered by POST /api/init has no effect; preflights from that origin stay 400",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5789",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F25",+ "feature_ids": [+ "F25.add-agent-server",+ "F25.health-status"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "known-issue",+ "reason": "Deferred-init server CORS origins delivered via/api/init are ignored by middleware; a real non-localhost Canvas origin then cannot admit/probe that backend. Requires that owned warm-pool server and actual non-localhost browser origin; the default loopback frontend is always allowed.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Deferred-init server CORS origins delivered via/api/init are ignored by middleware; a real non-localhost Canvas origin then cannot admit/probe that backend. Requires that owned warm-pool server and actual non-localhost browser origin; the default loopback frontend is always allowed.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5784,+ "title": "[Bug]: workspace session cookie ignores the server's path prefix, so prefixed workspace file URLs return 401",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5784",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F08",+ "feature_ids": [+ "F08.files-rich-plain",+ "F08.files-fallbacks",+ "F08.files-open-new-window"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "Path-prefixed local backend workspace previews use RemoteWorkspace.startWorkspaceSession and cookie authentication; the cookie path omits root_path and prefixed iframe/img/file requests return 401. Needs an authenticated prefixed server, real workspace asset and browser cookie readback. Surface owns the note, this ledger owns the row.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Path-prefixed local backend workspace previews use RemoteWorkspace.startWorkspaceSession and cookie authentication; the cookie path omits root_path and prefixed iframe/img/file requests return 401. Needs an authenticated prefixed server, real workspace asset and browser cookie readback. Surface owns the note, this ledger owns the row.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5740,+ "title": "[Bug]: Responses streaming crashes when OutputTextDeltaEvent lacks item_id",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5740",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "current-source-review",+ "reason": "Genuine Local report records SDK ConversationErrorEvent when Responses parser lacks item_id; current1.55 accesses event.item_id in display-delta processing.",+ "source_refs": [+ "src/contexts/conversation-websocket-context.tsx",+ "SDK1.55.0: openhands/sdk/llm/llm.py:_process_stream_event"+ ],+ "issue_state": "open",+ "fresh_live_outcome": null,+ "prerequisite": "Original incomplete SSE was not retained; genuine Responses stream/transformer path missing item_id required for UI replay. Pure parser reproducer proves SDK exception only.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5736,+ "title": "[Bug]: TypeScript hook regex alternatives match partial tool names",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5736",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Canvas does not call the TypeScript getHooksForEvent matching helper. Hook execution is server-side Python; Canvas only renders HookExecutionEvent matchers. The TS regex helper mismatch has no current executing Canvas consumer.",+ "source_refs": [+ "src/components/features/conversation-panel/hook-event-item.tsx",+ "src/components/features/conversation-panel/hook-matcher-content.tsx"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5731,+ "title": "Agent Server: whitespace-only session API keys still count as real keys",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5731",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F26",+ "feature_ids": [+ "F26.cli-public-needs-key"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "known-issue",+ "reason": "Canvas launcher forwards nonempty LOCAL_BACKEND_API_KEY without trimming; Agent Server config retains whitespace-only session keys. Recheck an owned loopback-only public deployment with a whitespace key, actual authentication and a normal-key control. The default generated strong key does not exercise it; SDK cipher fallback is distinct when Canvas supplies OH_SECRET_KEY.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Canvas launcher forwards nonempty LOCAL_BACKEND_API_KEY without trimming; Agent Server config retains whitespace-only session keys. Recheck an owned loopback-only public deployment with a whitespace key, actual authentication and a normal-key control. The default generated strong key does not exercise it; SDK cipher fallback is distinct when Canvas supplies OH_SECRET_KEY.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5729,+ "title": "[Bug]: PROVIDER_TOKEN_NAMES uses github_token instead of GITHUB_TOKEN",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5729",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F22",+ "feature_ids": [+ "F22.setup-repository-list"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md",+ "basis": "known-issue",+ "reason": "Automation 1.20 prompt/plugin preset workers call RemoteWorkspace.clone_repos, which reads lowercase provider names. An uppercase-only documented secret can leave an owned private repo unauthenticated. Needs a disposable authorized repository or reviewed token-gated local Git fixture through the real preset worker, plus a refused-loopback model configuration if initialization is needed; no paid inference is authorized.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Automation 1.20 prompt/plugin preset workers call RemoteWorkspace.clone_repos, which reads lowercase provider names. An uppercase-only documented secret can leave an owned private repo unauthenticated. Needs a disposable authorized repository or reviewed token-gated local Git fixture through the real preset worker, plus a refused-loopback model configuration if initialization is needed; no paid inference is authorized.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5718,+ "title": "[Bug]: Plugin or skill loading error should not prevent conversation from starting",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5718",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F19",+ "feature_ids": [+ "F19.enabled-autoload"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-plugins.tsx",+ "src/routes/launch.tsx",+ "src/components/features/plugins/",+ "src/components/features/launch/",+ "src/utils/plugin-launch-url.ts",+ "src/hooks/mutation/use-install-plugin.ts",+ "src/hooks/mutation/use-set-plugin-enabled.ts",+ "src/hooks/mutation/use-uninstall-plugin.ts",+ "src/hooks/mutation/use-refresh-plugin.ts"+ ],+ "changed_canvas_source_paths": [],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "SDK1.55/#5990 accepts valid root plugin.json without$schema; it does not establish recovery from INVALID JSON, which is the dated knownfailure fixture. Do not call that malformed-plugin case fixed from the layout fix."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 98+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5711,+ "title": "[Bug]: TypeScript RemoteState can retain stale snapshots and overwrite newer updates",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5711",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Canvas does not use TypeScript RemoteState; it owns event/state hydration through its frontend stores.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5652,+ "title": "[Bug]: RemoteState confirmation_policy/security_analyzer/agent fail or return None on the second read",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5652",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Python RemoteConversation cached getter behavior; Canvas consumes typed HTTP/WebSocket payloads rather than Python RemoteState.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5651,+ "title": "[Bug]: ConversationEventStream on Node 22 WebSocket stops after one refused reconnect instead of backing off",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5651",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Node 22 ConversationEventStream reconnect path; Canvas uses browser sockets and requires Node >=24 for its launcher.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5650,+ "title": "[Bug]: TS RemoteWorkspace.downloadAsBlob/fileDownload return 11 mangled bytes for a 5-byte binary file",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5650",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "RemoteWorkspace.downloadAsBlob/fileDownload are not Canvas download consumers; Canvas uses binary-safe FileClient.downloadFile.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5649,+ "title": "[Bug]: TS ConversationManager.createConversation ignores maxIterations/stuckDetection (server gets 500/true)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5649",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Canvas builds its create payload and calls ConversationClient directly, not TypeScript ConversationManager.createConversation.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5648,+ "title": "[Bug]: /app-backends WebSocket bridge negotiates no subprotocol (expected the backend's choice, qa.v1)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5648",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F20",+ "feature_ids": [+ "F20.app-backend-view"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "basis": "retained-historical-note",+ "reason": "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5647,+ "title": "[Bug]: Continued /v1/chat/completions turn reports conversation-total usage (13/7/20), not its own (3/2/5)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5647",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "OpenAI-compatible gateway per-turn usage contract; Canvas does not consume /v1/chat/completions gateway usage.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5645,+ "title": "[Bug]: Skill/Canvas install with an unknown ref returns 200 when the git source is cached (expected 400)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5645",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F19",+ "feature_ids": [+ "F19.add-plugin"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-plugins.tsx",+ "src/routes/launch.tsx",+ "src/components/features/plugins/",+ "src/components/features/launch/",+ "src/utils/plugin-launch-url.ts",+ "src/hooks/mutation/use-install-plugin.ts",+ "src/hooks/mutation/use-set-plugin-enabled.ts",+ "src/hooks/mutation/use-uninstall-plugin.ts",+ "src/hooks/mutation/use-refresh-plugin.ts"+ ],+ "changed_canvas_source_paths": [],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 94+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5644,+ "title": "[Bug]: /api/profiles/{name}/validate ignores provider_connection_id: linked draft valid:false, expected true",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5644",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "historical-only",+ "reason": "Canvas's LlmSettingsLocalView.handleSave explicitly skips ProfilesService.validateProfile when connectionId is set. The linked-draft validation request in this SDK report is not emitted by the mapped Canvas profile-save path.",+ "family": null,+ "map_file": null,+ "issue_state": "open",+ "previous_state": "out-of-scope",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5643,+ "title": "[Bug]: POST /api/settings/mcp/{key} with {} returns 201 and new conversations lose all MCP tools; expected 422",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5643",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F17",+ "feature_ids": [+ "F17.custom-add",+ "F17.agent-uses-server"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 106+ },+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 108+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5642,+ "title": "[Bug]: git changes returns 400 for a modified file with a space in its name and mangles non-ASCII names",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5642",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "live-reproduced",+ "reason": "Tracked my notes.md modification makes scoped git/changes return400 and hides changes.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F08.uncommitted-changes/sdk-5642-settled.png"+ ],+ "steps": [+ "In a disposable Git fixture, create and commit tracked my notes.md and caf\u00e9.txt; modify my notes.md.",+ "Open empty Local workspace conversation > right-panel-toggle > conversation-tab-commits.",+ "browser network --filter /git/ => conversation-scoped git/changes status400.",+ "Settled screenshot shows Uncommitted changes absent; only space-name case driven."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5639,+ "title": "[Bug]: LLM switched in before the first run is not added to ConversationStats; its tokens and cost are lost",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5639",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.llm-profile-picker"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A working model, a profile switch before the first run, and nonzero usage/cost after that turn.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model, a profile switch before the first run, and nonzero usage/cost after that turn.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5638,+ "title": "[Bug]: execute_bash_command returns only the last chunk of output over 1 MiB; null exit_code past 100 chunks",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5638",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-tree"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "live-reproduced",+ "reason": "A real1800-file listing over1MiB drops README.md and creates a partial-path bb folder before the2000-file cap.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F08.files-tree/sdk-5638-settled.png"+ ],+ "steps": [+ "In a disposable repository, arrange1800 small files beneath three220-character directory segments; total file count remains below2000 and sorted listing exceeds1MiB.",+ "Open its empty Local conversation > Files > files-tab-refresh.",+ "browser count role=button[name=\"README.md\"][exact] =>0 though real README.md remains.",+ "Files snapshot shows a spurious partial-path bb folder alongside surviving tail entries; retained screenshot records tree."+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5636,+ "title": "[Bug]: switch_llm returns 200 but keeps the old LLM when the posted usage_id is already registered",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5636",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "Canvas direct profile switches assign a new profile:<name>:<uuid> usage_id; the raw switch_llm default/already-registered usage_id collision is not its request shape.",+ "family": null,+ "map_file": null,+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "scope-reconfirmed",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5635,+ "title": "[Bug]: Deleting or sandbox-pausing a conversation leaves its open WebSockets silent instead of closing them",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5635",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.reconnect"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "live-reproduced",+ "reason": "prepare-for-sandbox-pause returned204 but the original communicating session socket stayed open for over10s. A newly opened Canvas tab received a new run:false message; the old tab did not. DELETE returned200 and GET404, but both held sockets stayed open over10s. Canvas later redirected through metadata polling; no claim of permanently stale navigation is made.",+ "verification": "live-reproduced",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "sdk": "1.54.0",+ "date": "2026-10-12",+ "browser": "Chromium153.0.8010.12",+ "steps": [+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser goto /conversations/<id>; POST a unique run:false user marker and wait for it; observe the communicating socket connectionId through browser events.",+ "control-openhands --run \"$OH_VERIFY_RUN\" api POST /api/conversations/prepare-for-sandbox-pause --write --data '{}'; observe socket metadata for at least10s without navigating/reloading the original tab.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser tab new /conversations/<id>; wait for original marker. POST QA_PAUSE_5635 as a run:false user event. New tab receives it; tab0 read-only eval document.body.innerText.includes(\"QA_PAUSE_5635\") was false.",+ "control-openhands --run \"$OH_VERIFY_RUN\" api DELETE /api/conversations/<id> --write; observe both original socket connection IDs for at least10s; GET /api/conversations/<id> returns404. Record later UI polling redirect separately.",+ "Save redacted browser events, both tab screenshots, and close/release only owned tabs."+ ],+ "actual": "prepare-for-sandbox-pause returned204 but the original communicating session socket stayed open for over10s. A newly opened Canvas tab received a new run:false message; the old tab did not. DELETE returned200 and GET404, but both held sockets stayed open over10s. Canvas later redirected through metadata polling; no claim of permanently stale navigation is made.",+ "artifacts": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-delete-sockets-5635.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-new-tab-5635.png",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F06.reconnect/pause-stale-tab-5635.png",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F06.reconnect/deleted-stale-tab-5635.png"+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5633,+ "title": "[Bug]: Failed run blocks the event loop while logging a rich traceback: /alive takes ~5 s, not ~100 ms",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5633",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.remote-backend"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Temporarily replaced only the owned Agent Server with the same pinned interpreter and state under genuine default Rich logging (LOG_JSON/CI/LOG_RICH_TRACEBACKS unset). Canvas Send uses tools=[] and a dummy LLM pointing at closed 127.0.0.1:9; 60 concurrent direct /alive probes had maximum latency 2.091 s while the conversation ended error. No provider reachable. Restored normal launcher and doctor afterward.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "live-reproduced",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "historical-reproduction",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5632,+ "title": "[Bug]: Waiting on a locked profile or connection store freezes the whole server; /alive stalls up to 30 s",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5632",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F11",+ "feature_ids": [+ "F11.list"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-row.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-actions-menu.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/delete-provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/llm-profiles-manager.tsx",+ "src/api/provider-connections-service/"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "SDK1.55/#5983 changes active-profile seeding, but does not offload blocking provider/profile store file locks. Existing25\u201328second responsiveness observations remain historical, not fresh current timings."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "line": 81+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5631,+ "title": "[Bug]: events/search and /count with kind=MessageEvent return nothing; only the full class path matches",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5631",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "SDK event-search short-kind alias contract; current Canvas event history requests do not send that short kind filter.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5629,+ "title": "[Bug]: A second server's first load of a shared conversation reverts meta.json to its stale startup copy",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5629",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.rename-inline"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "live-reproduced",+ "reason": "B was started with shared storage before A renamed an idle conversation from BEFORE_B to AFTER_B. A PATCH and disk confirmed AFTER_B. After own A graceful stop, native B events/count returned200 and rewrote disk title to its BEFORE_B startup copy. Real Add Backend\u2192B card showed BEFORE_B in Canvas. The native first-load request is explicit; an earlier sandbox-pause variant did not reach a valid first load and is not the claimed reproduction.",+ "verification": "live-reproduced",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "sdk": "1.54.0",+ "date": "2026-10-12",+ "browser": "Chromium153.0.8010.12",+ "steps": [+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "On A PATCH title QA_SHARED_BEFORE_B_5629_REPEAT. Start pinned SDK Python: python .agents/skills/verify-openhands/scripts/fixtures/shared-storage-server.py --run \"$OH_VERIFY_RUN\" --seconds 300 (background; no editor/frontend/browser; bounded owned PGID cleanup). It shares only this run's conversations/cipher/session key and records its loopback port in private/shared-server-5629/server.json.",+ "After B listens, A api PATCH /api/conversations/<id> --write --data '{\"title\":\"QA_SHARED_AFTER_B_5629_REPEAT\"}'; read only title/updated_at from shared meta.json and confirm AFTER_B.",+ "control-openhands --run \"$OH_VERIFY_RUN\" service stop agent-server; invoke pinned SDK Python shared-storage-server.py --run \"$OH_VERIFY_RUN\" --load-id <id> (native B events/count, observed200/count0). Immediately read shared meta.json: BEFORE_B was restored.",+ "control-openhands --run \"$OH_VERIFY_RUN\" restart; control-openhands --run \"$OH_VERIFY_RUN\" browser goto /?backend=default-local; control-openhands --run \"$OH_VERIFY_RUN\" doctor (A is restarted only to serve Canvas).",+ "Through real Add Backend select Agent-server, name QA_Shared_Repeat_5629, host=http://127.0.0.1:<B-port>, key via browser fill --value-file \"$OH_VERIFY_RUN\"/private/session-key, Connect, onboard --skip if needed, then click its BEFORE_B card. conversation-name-title displays BEFORE_B.",+ "Stop only the helper operator PID recorded in its own metadata (SIGTERM); verify its PGID/port gone. Return Canvas to default-local and eventually stop the owned primary."+ ],+ "actual": "B was started with shared storage before A renamed an idle conversation from BEFORE_B to AFTER_B. A PATCH and disk confirmed AFTER_B. After own A graceful stop, native B events/count returned200 and rewrote disk title to its BEFORE_B startup copy. Real Add Backend\u2192B card showed BEFORE_B in Canvas. The native first-load request is explicit; an earlier sandbox-pause variant did not reach a valid first load and is not the claimed reproduction.",+ "artifacts": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-repeat-before-b-load-5629.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-repeat-after-b-load-5629.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.rename-inline/shared-first-load-reverted-5629.png"+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5628,+ "title": "[Bug]: Resumed conversation keeps a since-disabled plugin's skills and commands; only its hooks are dropped",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5628",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F19",+ "feature_ids": [+ "F19.enabled-autoload"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-plugins.tsx",+ "src/routes/launch.tsx",+ "src/components/features/plugins/",+ "src/components/features/launch/",+ "src/utils/plugin-launch-url.ts",+ "src/hooks/mutation/use-install-plugin.ts",+ "src/hooks/mutation/use-set-plugin-enabled.ts",+ "src/hooks/mutation/use-uninstall-plugin.ts",+ "src/hooks/mutation/use-refresh-plugin.ts"+ ],+ "changed_canvas_source_paths": [],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 100+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5627,+ "title": "[Bug]: /goal loop active at a graceful restart stays `running` afterwards instead of `interrupted`",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5627",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.slash-goal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A working model goal loop active during graceful backend restart, then goal-status and resume readback.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model goal loop active during graceful backend restart, then goal-status and resume readback.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5626,+ "title": "[Bug]: Pause during a model call not saved: GET says running (expected paused); after SIGKILL it is error",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5626",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F06",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "live-not-reproduced",+ "reason": "During a genuine held dummy-loopback model request, Canvas Stop immediately persisted paused in REST and base_state.json. After SIGKILL of the verified own Agent Server listener and an owned restart, REST/disk and the reloaded Stopped UI retained paused. The bounded accept-only endpoint read no bytes and generated no model reply.",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/5626#issuecomment-6122465657",+ "steps": [+ "Start the tracked accept-only-endpoint.py fixture with its180s bound and a private output directory; read its port.json",+ "Create an idle tools-empty conversation with dummy key, loopback fixture base URL, timeout600 and autotitle false",+ "In Canvas submit a confined prompt; wait for Stop; confirm execution_status running and the fixture accepted a connection",+ "Click Stop; wait for Stopped; read execution_status paused from REST and owned base_state.json",+ "Identify the own Agent Server listener PID and prove ancestry includes this run launcher; SIGKILL only that PID",+ "control-openhands restart; doctor; read paused from REST/disk; reload and see Stopped; never press Play"+ ],+ "artifacts": [+ "evidence/F06.stop-resume/oct16-pause-durability.json",+ "evidence/F06.stop-resume/oct16-paused-after-crash.png"+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": "not-reproduced",+ "disposition": "parent-fresh-recheck-pending",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "current_result": "pass",+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Start the tracked accept-only-endpoint.py fixture with its180s bound and a private output directory; read its port.json",+ "Create an idle tools-empty conversation with dummy key, loopback fixture base URL, timeout600 and autotitle false",+ "In Canvas submit a confined prompt; wait for Stop; confirm execution_status running and the fixture accepted a connection",+ "Click Stop; wait for Stopped; read execution_status paused from REST and owned base_state.json",+ "Identify the own Agent Server listener PID and prove ancestry includes this run launcher; SIGKILL only that PID",+ "control-openhands restart; doctor; read paused from REST/disk; reload and see Stopped; never press Play"+ ],+ "observations": {+ "before_crash": {+ "api": "paused",+ "disk": "paused"+ },+ "after_restart": {+ "api": "paused",+ "disk": "paused"+ },+ "agent_server_listener_identity_verified": true,+ "ancestry_includes_owned_launcher": true,+ "accepted_connections": 1,+ "received_bytes_recorded": 0,+ "ui_after_restart": "Stopped"+ },+ "artifacts": [+ "evidence/F06.stop-resume/oct16-pause-durability.json",+ "evidence/F06.stop-resume/oct16-paused-after-crash.png"+ ],+ "no_successful_model_inference": true+ },+ "comment_id": 6122465657,+ "refresh_published_at": "2026-10-16T04:59:22Z",+ "body_sha256": "64cb28e803f5ab314d89690592348e96166a374618482699cad74c31cf7e5528",+ "run_id": "backfill-18334-settings-root",+ "needs_current_replay": false,+ "current_verdict": "not-reproduced",+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5624,+ "title": "[Bug]: Provider connection POST/PATCH accepts blank or redacted api_key values",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5624",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F11",+ "feature_ids": [+ "F11.create-validation",+ "F11.edit"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-row.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-actions-menu.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/delete-provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/llm-profiles-manager.tsx",+ "src/api/provider-connections-service/"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "line": 79+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5622,+ "title": "[Bug]: After a cipher key change, settings and secret writes answer 200 and null stored keys, not 409",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5622",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F14",+ "feature_ids": [+ "F14.create",+ "F14.edit",+ "F14.delete"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F14-secrets.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/secrets-settings.tsx",+ "src/components/features/settings/secrets-settings/",+ "src/api/secrets-service.ts",+ "src/components/features/conversation/secret-request-modal.tsx",+ "src/services/secret-request.ts",+ "src/stores/secret-request-store.ts",+ "src/constants/secret-request.ts",+ "src/api/request-secret-client-tool.ts"+ ],+ "changed_canvas_source_paths": [+ "src/api/request-secret-client-tool.ts",+ "src/components/features/conversation/secret-request-modal.tsx",+ "src/components/features/settings/secrets-settings/secret-form.tsx",+ "src/constants/secret-request.ts",+ "src/services/secret-request.ts",+ "src/stores/secret-request-store.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Cipher mismatch write refusal56f4777f5 and keyrotation46fbbccb6 are NOT in pinnedSDKv1.55.0. Preserve the dated mismatch observations; no Oct16 cipher replacement was performed."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F14-secrets.md",+ "line": 69+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5607,+ "title": "[Bug]: POST /api/mcp/test classifies a remote MCP server's HTTP 401 as error_kind \"unknown\", so Agent Canvas tells a user with a wrong GitHub token \"Connection failed\"",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5607",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F17",+ "feature_ids": [+ "F17.credential-probe"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 42+ },+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 133+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5590,+ "title": "[Bug]: api_lifespan ignores the Config passed to create_app(), leaking conversations into OH_CONVERSATIONS_PATH",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5590",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Python create_app(Config) lifespan embedding contract; Canvas launches agent-server using its CLI/environment configuration.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5539,+ "title": "[Bug]: download-trajectory fails or returns a broken zip when the same trajectory is downloaded again or concurrently",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5539",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.download-zip"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "live-reproduced",+ "reason": "A20-second three-client real loopback overlapping receiver made929 attempts:31 completed corrupt200 ZIPs,106HTTP500,269ingress502,133RemoteProtocolError,388 validZIPs and2 intentionally deadline-aborted reads. Both ambiguous Content-Length differences were excluded from the public claim. Actual Canvas Download during overlap produced a ZIP with inspectable entries; that particular UI download is not claimed corrupt. A separate sequential2000-request pass returned validZIPs throughout.",+ "verification": "live-reproduced",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "sdk": "1.54.0",+ "date": "2026-10-12",+ "browser": "Chromium153.0.8010.12",+ "steps": [+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "Use the pinned SDK Python (httpx installed): python .agents/skills/verify-openhands/scripts/fixtures/download-trajectory-overlap.py --base-url <own-base-url> --conversation-id <id> --api-key-file \"$OH_VERIFY_RUN\"/private/session-key --output-dir \"$OH_VERIFY_RUN\"/private/trajectory-overlap --seconds 20 --clients 3 --rounds 5000 --slow-ms 2 (background). It rejects non-loopback origins, caps network/uncompressed bytes/members, checks CRC in bounded chunks, never treats aborted reads as broken, and closes every client.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser goto /conversations/<id>; control-openhands --run \"$OH_VERIFY_RUN\" browser click \"testid=chat-pane-header >> testid=ellipsis-button\"; control-openhands --run \"$OH_VERIFY_RUN\" browser click testid=download-trajectory-button while real receivers overlap.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser downloads --last 3 --inspect; retain the real receiver sanitized summary. ZIP samples and session keys remain private.",+ "Read the public corrected summary:31 completed brokenZIPs supports the race;2 original length differences were ambiguous with2 deadline aborts, so no length/truncation count is asserted."+ ],+ "actual": "A 20-second three-client real loopback overlapping receiver made929 attempts:31 completed corrupt200 ZIPs,106HTTP500,269ingress502,133RemoteProtocolError,388 validZIPs and2 intentionally deadline-aborted reads. Both ambiguous Content-Length differences were excluded from the public claim. Actual Canvas Download during overlap produced a ZIP with inspectable entries; that particular UI download is not claimed corrupt. A separate sequential2000-request pass returned validZIPs throughout.",+ "artifacts": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.download-zip/trajectory-overlap-5539-summary.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.download-zip/trajectory-summary-correction.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.download-zip/overlapping-download-5539.png"+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5533,+ "title": "Tmux can accept commands before slow Bash startup completes",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5533",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "A model key for real agent-driven first-command observations is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A model key for real agent-driven first-command observations is unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5523,+ "title": "[Bug]: deepseek-v4.1-flash is not detected as vision-capable (supports_vision=False), so its profile is excluded from the vision fallback",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5523",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.image-only-send"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A working DeepSeek V4.1 Flash vision profile and another fallback profile, with an image-bearing turn.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working DeepSeek V4.1 Flash vision profile and another fallback profile, with an image-bearing turn.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5501,+ "title": "[Bug]: Canvas extension manifest drops pages[].nav_label, so Agent Canvas shows the page title in the rail",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5501",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F20",+ "feature_ids": [+ "F20.rail-label"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "basis": "retained-historical-note",+ "reason": "Existing Agent Server nav_label omission note and F20.rail-label recipe preserved.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "existing-note",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "not-run",+ "verification_detail": "historical-only-no-current-replay"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5498,+ "title": "[Bug]: After deleting the active LLM profile, its provider connection cannot be deleted (\"referenced by the active agent settings\")",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5498",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F11",+ "feature_ids": [+ "F11.delete-stale-reference"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-row.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-actions-menu.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/delete-provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/llm-profiles-manager.tsx",+ "src/api/provider-connections-service/"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "line": 93+ }+ ],+ "map_note_present": true,+ "verification_status": "not-run"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5497,+ "title": "[Bug]: Rotating a provider connection's API key does not reach the active LLM settings; new conversations keep using the old key",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5497",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F11",+ "feature_ids": [+ "F11.rotate"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-row.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-actions-menu.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/delete-provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/llm-profiles-manager.tsx",+ "src/api/provider-connections-service/"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Provider-key cascade40021e951 is NOT in SDKv1.55.0, so the current pin does not supply that proposed repair."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "line": 99+ }+ ],+ "map_note_present": true,+ "prerequisite": "See the issue-specific prerequisite in the linked Known issue note.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5496,+ "title": "[Bug]: Concurrent plugin requests corrupt .installed.json, so a just-installed plugin is recorded as source \"local\" and Update/launch then fail",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5496",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F19",+ "feature_ids": [+ "F19.refresh",+ "F19.start-conversation",+ "F19.enabled-autoload"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-plugins.tsx",+ "src/routes/launch.tsx",+ "src/components/features/plugins/",+ "src/components/features/launch/",+ "src/utils/plugin-launch-url.ts",+ "src/hooks/mutation/use-install-plugin.ts",+ "src/hooks/mutation/use-set-plugin-enabled.ts",+ "src/hooks/mutation/use-uninstall-plugin.ts",+ "src/hooks/mutation/use-refresh-plugin.ts"+ ],+ "changed_canvas_source_paths": [],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Shared installation metadata lock PR5690 is not a shippedv1.55.0 commit. New5985 describes a separate directory-swap/list enabled-state race; keep the original concurrent metadata note distinct."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 71+ },+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 126+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5495,+ "title": "[Bug]: condenser_kind is exposed as a user-facing settings field with a reStructuredText developer description",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5495",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F15",+ "feature_ids": [+ "F15.condenser-kind"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/condenser-settings.tsx",+ "src/routes/agent-context-settings.tsx",+ "src/routes/verification-settings.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/utils/sdk-settings-schema.ts"+ ],+ "changed_canvas_source_paths": [+ "src/utils/sdk-settings-schema.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "line": 98+ }+ ],+ "map_note_present": true,+ "verification_status": "not-run"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5492,+ "title": "[Bug]: goal judge treats `\"complete\": \"false\"` (a string) as complete",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5492",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.slash-goal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A real goal judge response containing complete as the string false rather than a JSON boolean.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real goal judge response containing complete as the string false rather than a JSON boolean.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5475,+ "title": "[Bug]: version-bump-prs has never opened the enterprise SDK bump PR (403); land #5339",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5475",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Enterprise SDK dependency-bump CI permissions; no Canvas feature or user flow.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5466,+ "title": "[Bug]: A stale codex-acp / claude-agent-acp on PATH silently replaces the pinned ACP version (Codex ACP fails to start)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5466",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "PinnedACP adapter catalog remains relevant; the old PATH-selected0.16.0 preauthentication observation does not prove a working account/session today."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 87+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5465,+ "title": "[Bug]: A forked conversation loses its LLM API key and fails its first turn with LLMAuthenticationError",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5465",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.branch-from-here"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working cipher-backed model conversation, then fork and a successful turn on the fork.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working cipher-backed model conversation, then fork and a successful turn on the fork.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5447,+ "title": "[Bug]: claude-opus-5-5 requests ending in an assistant (prefill) message fail with Anthropic 400 \"does not support assistant message prefill\"",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5447",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "Anthropic Claude Opus5.5 credentials and a real trailing-assistant-prefill request.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Anthropic Claude Opus5.5 credentials and a real trailing-assistant-prefill request.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5375,+ "title": "[Bug]: GitLab MCP OAuth fails with localhost redirect URI",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5375",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F17",+ "feature_ids": [+ "F17.install-oauth",+ "F17.agent-uses-server"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "line": 140+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5373,+ "title": "agent-server: a logging call can wedge the event loop and amplify memory (rich traceback render)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5373",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.remote-backend"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "A separately hosted default-Rich backend and the original cyclic-exception peer/backlog driver are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A separately hosted default-Rich backend and the original cyclic-exception peer/backlog driver are unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5354,+ "title": "[Bug]: Validation errors split a single tool-call response in SDK conversation history",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5354",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.event-groups"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A live model turn mixing valid, invalid and valid tool calls in one response.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A live model turn mixing valid, invalid and valid tool calls in one response.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5330,+ "title": "[Bug]: /btw causes prompt-cache invalidation",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5330",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.slash-btw"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A working caching provider and comparable ordinary-message versus /btw prompt/cache records.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working caching provider and comparable ordinary-message versus /btw prompt/cache records.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5322,+ "title": "[Bug]: Pager guard from #3660/#3684 doesn't cover TmuxPanePool or WindowsTerminal",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5322",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "A model key for genuine terminal-tool execution is unavailable; the PowerShell half additionally requires Windows.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A model key for genuine terminal-tool execution is unavailable; the PowerShell half additionally requires Windows."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5310,+ "title": "[Bug]: Docker conversation runtime cannot resolve configured title LLM profile because per-runtime profile store is empty",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5310",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.docker-conversation-runtime"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Docker daemon/per-conversation image and real auxiliary-title/ACP credentials are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Docker daemon/per-conversation image and real auxiliary-title/ACP credentials are unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5307,+ "title": "[Bug]: Docker conversation runtime creates linked worktree across incompatible host/container filesystem namespaces",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5307",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.docker-conversation-runtime"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "Docker daemon/per-conversation image is unavailable; a host/container git fixture is required.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Docker daemon/per-conversation image is unavailable; a host/container git fixture is required."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5279,+ "title": "fix(workspace): link attached conversations in automation callbacks",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5279",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.run-linked-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "An authenticated Cloud automation workspace/callback context and owned stable conversation are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "An authenticated Cloud automation workspace/callback context and owned stable conversation are unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5252,+ "title": "[Testing gap]: No cross-version conversation fixture, so a dependency pin can make older persisted conversations unreadable without CI noticing",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5252",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Cross-version test-fixture gap; no reported user-visible failure to reproduce.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5225,+ "title": "[Bug]: LLMBadRequestError from a poisoned tool_result permanently wedges conversations \u2014 no recovery path in Agent.step",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5225",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A real provider-rejected poisoned image tool_result in a disposable persisted history, followed by retry.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real provider-rejected poisoned image tool_result in a disposable persisted history, followed by retry.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5207,+ "title": "OpenRouter: resolve shared connections across runtime paths and preserve structured reasoning",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5207",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F11",+ "feature_ids": [+ "F11.agent-uses-connection",+ "F11.rotate"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-row.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-actions-menu.tsx",+ "src/components/features/settings/llm-profiles/provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/delete-provider-connection-modal.tsx",+ "src/components/features/settings/llm-profiles/llm-profiles-manager.tsx",+ "src/api/provider-connections-service/"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md",+ "line": 100+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5181,+ "title": "[Bug]: output-cap defaults and explicit reasoning overrides are mishandled",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5181",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.schema-validation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 172+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5167,+ "title": "[Bug]: Codex ACP on a headless host: `authenticate(\"chat-gpt\")` does not complete within `ACP_AUTH_TIMEOUT` although Codex is logged in (\"ChatGPT authentication did not complete in time\")",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5167",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 129+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5152,+ "title": "Unlocked installs resolve browser-use 0.11.13 with mcp 2.x \u2014 browser tools silently disappear",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5152",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F27",+ "feature_ids": [+ "F27.browser-screenshot"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "reason": "Using the actual owned Agent Server 1.55.0 interpreter and private HOME, BrowserToolExecutor() constructs successfully. The resolved browser-use 0.11.13, mcp 1.30.0 and fastmcp 3.4.8 do not hit the reported mcp 2.x import failure. This is the SDK report\u2019s explicit constructor reproduction; it does not establish a successful Canvas browser operation.",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/5152#issuecomment-6121793764",+ "basis": "live-check",+ "issue_state": "open",+ "previous_outcome": "not-reproduced",+ "historical_basis": null,+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "pass",+ "fresh_reproduction_claim": false,+ "disposition": "current-negative-rechecked",+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "steps": [+ "Identify only the owned launcher descendant interpreter: /Users/enyst/.cache/uv/archive-v0/pckcVACfXtx70WBA/bin/python. Set HOME to the run\u2019s private/home and retain closed-loopback OpenAI fallback/AWS metadata protections.",+ "Run that interpreter with: from openhands.tools.browser_use.impl import BrowserToolExecutor; x=BrowserToolExecutor(). Read importlib.metadata.version for openhands-agent-server, openhands-sdk, browser-use, mcp and fastmcp. No browser operation or model request."+ ],+ "artifacts": [+ "evidence/F27.browser-screenshot/current-sdk5152-constructor.json"+ ],+ "verification_status": "not-reproduced",+ "comment_id": 6121793764,+ "refresh_published_at": "2026-10-16T04:40:20Z",+ "body_sha256": "e11786a76700cc5f7e6beabbec6adba082eb96345d5e2bcd0ceec7f62d3d70dd"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5138,+ "title": "Cloud-proxy (TS client) sends upstream body as object -> 422 from /api/cloud-proxy (Pydantic rejects non-string body)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5138",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Retired TypeScript cloud-proxy body API; current Canvas makes direct Cloud App-API calls and uses typed runtime clients.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5108,+ "title": "[Bug]: Gemini edit corrupts non-UTF-8 content during unrelated edits",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5108",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "An Agent Server configured with the native Gemini edit tool set and a real tool edit of an invalid-UTF8 fixture; Gemini ACP uses a different external executor.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An Agent Server configured with the native Gemini edit tool set and a real tool edit of an invalid-UTF8 fixture; Gemini ACP uses a different external executor.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5100,+ "title": "[Bug]: Profile pre-flight validation reports budget-exceeded keys as valid and blocks ~2 min on retries",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5100",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.create-validation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 110+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5096,+ "title": "[Bug]: FileEditor replaces symlinks instead of updating the target file",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5096",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working model invoking canonical file-editor str_replace/insert through a real symlink, then undo.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model invoking canonical file-editor str_replace/insert through a real symlink, then undo.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5094,+ "title": "[Bug]: ACP session/load failure silently starts a fresh session without restoring conversation context",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5094",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.open-by-url"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "An authenticated Codex ACP session whose original session/load fails while persisted history remains.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Codex ACP session whose original session/load fails while persisted history remains.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5092,+ "title": "WAITING_FOR_CONFIRMATION: any user message implicitly approves pending actions on next run()",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5092",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.confirmation-mode"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working model proposing a real confirmation-gated marker action, then a revocation user message.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model proposing a real confirmation-gated marker action, then a revocation user message.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5074,+ "title": "TaskTool child remains running and task mapping is lost after Agent Server process restart",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5074",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A native delegated child still running during an owned server crash and parent recovery.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A native delegated child still running during an owned server crash and parent recovery.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5038,+ "title": "[Bug]: UTF-8 Chinese text misclassified as binary when sampling splits a multibyte character",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5038",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A live canonical file-editor view/edit on the specific UTF8 sample alignment; Canvas Files preview uses a separate NUL sniff, not this validator.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A live canonical file-editor view/edit on the specific UTF8 sample alignment; Canvas Files preview uses a separate NUL sniff, not this validator.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5027,+ "title": "fix: preserve repository identity in PR linked-issue readiness checks",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5027",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "PR linked-issue readiness-check repository identity; repository CI tooling.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4990,+ "title": "[Bug]: /pause reports PAUSED while an ACP agent keeps executing tool calls",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4990",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "An authenticated Claude Code ACP turn executing separate marker-writing tools while Pause is pressed.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated Claude Code ACP turn executing separate marker-writing tools while Pause is pressed.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4987,+ "title": "[Bug]: Title generation prompt glues user message to instruction and breaks emoji list formatting",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4987",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "known-issue",+ "reason": "A working autotitle model and captured title-generation input/output; Canvas only consumes the resulting server title.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working autotitle model and captured title-generation input/output; Canvas only consumes the resulting server title.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4983,+ "title": "[Bug]: system grep fallbacks misinterpret extended regular expressions",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4983",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A runtime without ripgrep and a live agent grep call using extended regex and escaped literal punctuation.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A runtime without ripgrep and a live agent grep call using extended regex and escaped literal punctuation.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4965,+ "title": "[Bug]: to_chat_dict emits blank text blocks and empty content arrays that strict OpenAI-compatible providers reject",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4965",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A strict OpenAI-compatible provider and a real history containing blank text or empty content arrays.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A strict OpenAI-compatible provider and a real history containing blank text or empty content arrays.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4963,+ "title": "[Bug]: Windows PowerShell terminal is permanently poisoned after a terminating error or interrupt",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4963",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "basis": "retained-historical-note",+ "reason": "Native Windows PowerShell terminal runtime is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Native Windows PowerShell terminal runtime is unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4934,+ "title": "SDK sends unsupported reasoning_effort to MiniMax M3",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4934",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.schema-validation",+ "F10.default-used"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 173+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4912,+ "title": "Path-gated test jobs omit openhands-sdk/**, so SDK changes can break downstream tests with every required check green",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4912",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "SDK CI path-gating and downstream test selection; no Canvas runtime feature.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4901,+ "title": "ACP subprocess leaf process orphaned on shutdown (POSIX) \u2014 process-group teardown gap, separate from #4409's Windows fix",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4901",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "An owned authenticated ACP provider runtime/command chain is unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "An owned authenticated ACP provider runtime/command chain is unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4893,+ "title": "agent-server: conversation stuck in execution_status=RUNNING after sandbox is paused/stopped mid-run (unresumable via API)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4893",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.status-menu"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "known-issue",+ "reason": "A disposable sandbox/VM frozen mid-run with a live model and persisted ownership state.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A disposable sandbox/VM frozen mid-run with a live model and persisted ownership state.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4891,+ "title": "MCP connection failure aborts send_message before the LLM is called",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4891",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F17",+ "feature_ids": [+ "F17.agent-uses-server"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "live-not-reproduced",+ "reason": "Two real initializations with a missing stdio executable and refused HTTP sibling retained the healthy qa_echo tool, persisted user messages and reached the expected loopback model error. Successful provider replies remain outside this check.",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4891#issuecomment-6121790068",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-reproduced",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @/private/tmp/odie-18309-QA_LOOPBACK.json",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands fixture mcp-server --name qa-settings-echo",+ "control-openhands api POST /api/settings/mcp/qa_settings_echo --write --data @<echo-config.json>",+ "control-openhands browser goto /mcp",+ "control-openhands browser click 'testid=mcp-health-probe-qa_settings_echo' --observe 'testid=mcp-health-label-qa_settings_echo' --observe-ms 6000",+ "control-openhands browser text 'testid=mcp-health-label-qa_settings_echo'",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation events 063a0efe-1766-457f-a27f-11edbb9e4345 --kinds SystemPromptEvent --from-start --last 1",+ "control-openhands api POST /api/settings/mcp/qa_settings_bad --write --data '{\"transport\":\"stdio\",\"command\":\"/qa/no-such-binary\",\"enabled\":true}'",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation events 701d69e9-b476-4170-a165-0742aae18196 --kinds SystemPromptEvent --from-start --last 1",+ "control-openhands conversation status 701d69e9-b476-4170-a165-0742aae18196",+ "control-openhands api DELETE /api/settings/mcp/qa_settings_bad --write",+ "control-openhands api POST /api/settings/mcp/qa_unreachable --write --data '{\"transport\":\"streamable-http\",\"url\":\"http://127.0.0.1:1/mcp\",\"enabled\":true}'",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation events 2653fd2e-6cea-48f9-a15e-a8351fba3cda --kinds SystemPromptEvent --from-start --last 1",+ "control-openhands conversation status 2653fd2e-6cea-48f9-a15e-a8351fba3cda",+ "control-openhands api DELETE /api/settings/mcp/qa_unreachable --write",+ "control-openhands api DELETE /api/settings/mcp/qa_settings_echo --write",+ "control-openhands doctor"+ ],+ "evidence": [],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Missing stdio and unreachable HTTP MCP sources both allowed initialization and preserved the healthy sibling tool, then reached the expected loopback model error. No successful reply was claimed.",+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "/private/tmp/odie-18309-QA_LOOPBACK.json",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "api_key": "<dummy value read from private dummy-key file>",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ },+ "echo_config": {+ "file": "<echo-config.json>",+ "body": {+ "transport": "stdio",+ "command": "<Node executable returned by fixture mcp-server>",+ "args": [+ "<server.mjs returned by fixture mcp-server>"+ ],+ "enabled": true+ }+ }+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "SDK1.55/#4838 changes MCP HTTP session recovery; previous graceful startup-failure observations are historical until root currentreplay."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Arrange one real qa_echo stdio fixture, one missing executable and one refused127.0.0.1:1 HTTP MCP server",+ "Two control-openhands conversation start UI launches with confined no-tools prompts",+ "Wait for error; inspect SystemPromptEvent tools and persisted MessageEvent",+ "Remove only faulty own MCP fixtures"+ ],+ "observations": {+ "healthy_tool_retained_in_both": true,+ "user_message_persisted_in_both": true,+ "reached_loopback_model_error": true+ },+ "artifacts": [],+ "no_successful_model_inference": true+ },+ "current_result": "pass",+ "fresh_live_outcome": "not-reproduced",+ "comment_id": 6121790068,+ "refresh_published_at": "2026-10-16T04:40:22Z",+ "body_sha256": "8e11201ae4d84af01fab2d0b731a067498db7643da50810088a3835e8c0b1de3",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "Arrange one real qa_echo stdio fixture, one missing executable and one refused127.0.0.1:1 HTTP MCP server",+ "Two control-openhands conversation start UI launches with confined no-tools prompts",+ "Wait for error; inspect SystemPromptEvent tools and persisted MessageEvent",+ "Remove only faulty own MCP fixtures"+ ],+ "artifacts": [],+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4847,+ "title": "[Bug]: RemoteConversation.fork() ignores title in server response; test mocks wrong server shape",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4847",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Python RemoteConversation.fork() wrapper title parsing; Canvas calls the typed REST fork endpoint directly.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4846,+ "title": "[Bug]: WebSocket clients ignore Node's global WebSocket in ESM builds",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4846",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Node ESM WebSocket constructor selection; Canvas browser selects window.WebSocket.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4811,+ "title": "[Bug]: fork sets title by injecting into state.tags instead of a dedicated field",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4811",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "The report is about standalone SDK fork title injection into ConversationState.tags; it explicitly separates Agent Server StoredConversation title handling used by Canvas.",+ "family": null,+ "map_file": null,+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "scope-reconfirmed",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4810,+ "title": "[Bug]: confirmation_policy, security_analyzer, and secrets mutations not persisted to meta.json",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4810",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "Canvas sets policy/analyzer/secrets at creation or through account settings/resume secrets; it does not call these per-conversation mutation endpoints.",+ "family": null,+ "map_file": null,+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "scope-reconfirmed",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4800,+ "title": "[Bug]: ConversationService: stored fields mutated in memory before save_meta() with no rollback on failure",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4800",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F07",+ "feature_ids": [+ "F07.rename-inline"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "basis": "live-reproduced",+ "reason": "After a successful baseline UI rename, chmod0500 on only the owned conversation directory made the next actual UI rename fail with PermissionError/HTTP500. GET nevertheless reported the rejected title while meta.json retained the baseline. Permissions were restored; graceful restart then persisted the rejected in-memory title. Tags/autotitle branches were not tested.",+ "verification": "live-reproduced",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "sdk": "1.54.0",+ "date": "2026-10-12",+ "browser": "Chromium153.0.8010.12",+ "steps": [+ "Launch this checkout at fa1f724ab5737a17453ad6081129eb91c5a423c6 in a new isolated run, then control-openhands doctor --run \"$OH_VERIFY_RUN\". SDK 1.54.0, automation1.19.3, Chromium 153.0.8010.12.",+ "Use a dummy profile with model=openai/qa-rechecks, api_key=qa-not-a-real-key, base_url=http://127.0.0.1:9, num_retries=0, timeout=2; autotitle=false. No provider inference is performed.",+ "Create an idle conversation via api POST /api/conversations --write --data @<idle-json>, with agent_settings.agent_kind=openhands, llm=<dummy-llm>, tools=[], LocalWorkspace working_dir=<owned-workspace>, autotitle=false, and no initial_message.",+ "control-openhands --run \"$OH_VERIFY_RUN\" browser goto /conversations/<id>; dblclick testid=conversation-name-title, fill conversation-name-input QA_META_BEFORE_4800, press Enter; verify GET title and disk meta title.",+ "Record the original permissions of \"$OH_VERIFY_RUN\"/private/state/dev_conversations/<id-hex>, then chmod 0500 only that owned directory (a read-only meta file alone cannot block an atomic replacement).",+ "Through the same UI rename to QA_META_REFUSED_4800. Observe permission-denied toast/HTTP500; GET title is REFUSED, but read only title/updated_at from meta.json and observe BEFORE.",+ "Restore the directory exact original mode in finally. control-openhands --run \"$OH_VERIFY_RUN\" restart; control-openhands --run \"$OH_VERIFY_RUN\" doctor; GET/reload and disk now show the previously rejected REFUSED title persisted during graceful shutdown.",+ "Retain only sanitized title/time snapshots; no disk-full/global filesystem changes or raw stored secrets."+ ],+ "actual": "After a successful baseline UI rename, chmod0500 on only the owned conversation directory made the next actual UI rename fail with PermissionError/HTTP500. GET nevertheless reported the rejected title while meta.json retained the baseline. Permissions were restored; graceful restart then persisted the rejected in-memory title. Tags/autotitle branches were not tested.",+ "artifacts": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-before-4800.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-denied-4800.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.rename-inline/refused-title-4800.png",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-rechecks/evidence/F07.rename-inline/metadata-after-graceful-restart-4800.json"+ ],+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4777,+ "title": "[Bug]: ParallelToolExecutor starts a pending tool after cancellation while waiting for a resource lock",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4777",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.stop-resume"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A live batch with a second real tool waiting on a declared-resource lock when Stop is pressed.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A live batch with a second real tool waiting on a declared-resource lock when Stop is pressed.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4776,+ "title": "[Bug]: MCPToolObservation silently discards structuredContent and result _meta",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4776",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "historical-only",+ "reason": "There is no structuredContent, structured_content, result_meta or resultMeta consumer under current src/. The report demonstrates SDK conversion/persistence metadata loss, without a current mapped Canvas renderer or API consumer for those fields.",+ "family": null,+ "map_file": null,+ "issue_state": "open",+ "previous_state": "out-of-scope",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4774,+ "title": "[Bug]: MessageEvent.visualize returns blank output for an empty TextContent block",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4774",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Python MessageEvent.visualize() CLI terminal rendering; Canvas renders MessageEvent content in its own React components.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4765,+ "title": "[Bug]: redact_text_secrets leaves api_key assignments with spaces around = unredacted",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4765",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "historical-only",+ "reason": "The affected redact_text_secrets helper is used by SDK command and ACP diagnostic logging. This report supplies only a direct Python helper call and no affected Canvas-rendered event, endpoint field or user control; F14 secret CRUD does not call it.",+ "family": null,+ "map_file": null,+ "issue_state": "open",+ "previous_state": "out-of-scope",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4731,+ "title": "Fix WebSocket client require('ws') breaking bundlers",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4731",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "TypeScript package bundler compatibility for require(ws); current Canvas browser event transport is separate and the frozen Canvas production build succeeds.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4709,+ "title": "[Bug]: AgentContext.current_datetime persists in settings.json, causing stale CURRENT_DATETIME in prompts",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4709",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F15",+ "feature_ids": [+ "F15.agent-context-datetime"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md",+ "basis": "live-not-reproduced",+ "reason": "Two UI launches with an explicit2001 settings timestamp produced fresh runtime datetimes and no2001 timestamp in events. Two natural-stale controls also used fresh datetimes. The original setting was restored.",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4709#issuecomment-6121790419",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-reproduced",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @/private/tmp/odie-18309-QA_LOOPBACK.json",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands api GET /api/settings --pick agent_settings.agent_context.current_datetime",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "python3 - <<'PY_DATE'\nimport json, os\nfrom pathlib import Path\nfor p in (Path(os.environ['OH_VERIFY_RUN']) / 'private').rglob('base_state.json'):\n d=json.loads(p.read_text())\n print({'conversation':p.parent.name,'runtime_datetime':d.get('agent',{}).get('agent_context',{}).get('current_datetime')})\nPY_DATE",+ "control-openhands api PATCH /api/settings --write --data '{\"agent_settings_diff\":{\"agent_context\":{\"current_datetime\":\"2001-01-01T00:00:00+00:00\"}}}' --pick agent_settings.agent_context.current_datetime",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation events 950e5e83-f56d-4dbb-9aab-b2ccc78f1972 --kinds SystemPromptEvent --grep 2001-01-01 --from-start --last 1",+ "python3 - <<'PY_DATE'\nimport json, os\nfrom pathlib import Path\nfor p in (Path(os.environ['OH_VERIFY_RUN']) / 'private').rglob('base_state.json'):\n d=json.loads(p.read_text())\n print({'conversation':p.parent.name,'runtime_datetime':d.get('agent',{}).get('agent_context',{}).get('current_datetime')})\nPY_DATE",+ "control-openhands api PATCH /api/settings --write --data '{\"agent_settings_diff\":{\"agent_context\":{\"current_datetime\":\"2026-10-12T07:29:16.188868+02:00\"}}}' --pick agent_settings.agent_context.current_datetime",+ "control-openhands doctor"+ ],+ "evidence": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker/evidence/F15.agent-context-page/datetime-diagnostic.json"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Canvas launches used fresh runtime datetimes despite naturally stale settings and an explicit old-date fixture. The old date was absent from the initialized prompt.",+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "/private/tmp/odie-18309-QA_LOOPBACK.json",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "api_key": "<dummy value read from private dummy-key file>",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ },+ "datetime_restore": "Capture the original settings timestamp with the first GET, then substitute it into the final PATCH. The literal 2026-10-12 value was the observed original in this run."+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/condenser-settings.tsx",+ "src/routes/agent-context-settings.tsx",+ "src/routes/verification-settings.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/utils/sdk-settings-schema.ts"+ ],+ "changed_canvas_source_paths": [+ "src/utils/sdk-settings-schema.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "No current completed datetime check; old explicitstale timestamp and userprompt diagnostics stay datedOct12. Root owns currentfixture replay."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Arrange QA_ROOT_BASE dummy profile with base_url127.0.0.1:1/v1; activate the seeded default OpenHands agent",+ "control-openhands fixture skill --name qa-root-legacy --trigger qa-root-legacy-ping",+ "Move only this run-owned skill into <run>/private/skills",+ "control-openhands browser goto /skills",+ "control-openhands conversation start --prompt \"qa-root-legacy-ping. Reply with exactly pong. Do not run any tools.\"",+ "For datetime case: PATCH agent_settings_diff.agent_context.current_datetime=2001-01-01; repeat two UI starts; restore original",+ "Read owned base_state/event files; emit only key-present, datetime and skill-match diagnostics"+ ],+ "observations": {+ "initializations": [+ {+ "conversation": "2c860cb9-1352-4a51-957f-20641757b938",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T04:24:42.885562Z",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "9ee3edc4-d55c-42fc-b778-2b288c68686b",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T04:24:44.293887Z",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "546e89fc-f933-46d7-9d27-c6d1a3c946c9",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T06:23:21.923984+02:00",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "415433ba-f805-4da2-bebd-f6fa43f6e081",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T06:23:20.216502+02:00",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ }+ ]+ },+ "artifacts": [],+ "no_successful_model_inference": true+ },+ "current_result": "pass",+ "fresh_live_outcome": "not-reproduced",+ "comment_id": 6121790419,+ "refresh_published_at": "2026-10-16T04:40:23Z",+ "body_sha256": "90c89c378e08f9648c87e145b2a1b9c5afbed0ae0f0edf6b9a898bf0250e72fd",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "Arrange QA_ROOT_BASE dummy profile with base_url127.0.0.1:1/v1; activate the seeded default OpenHands agent",+ "control-openhands fixture skill --name qa-root-legacy --trigger qa-root-legacy-ping",+ "Move only this run-owned skill into <run>/private/skills",+ "control-openhands browser goto /skills",+ "control-openhands conversation start --prompt \"qa-root-legacy-ping. Reply with exactly pong. Do not run any tools.\"",+ "For datetime case: PATCH agent_settings_diff.agent_context.current_datetime=2001-01-01; repeat two UI starts; restore original",+ "Read owned base_state/event files; emit only key-present, datetime and skill-match diagnostics"+ ],+ "artifacts": [],+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4705,+ "title": "[Bug]: Missing GitHub MCP inputSchema causes inconsistent task status between the task list and automation page",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4705",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling",+ "F23.run-logs-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Affected client/server versions, GitHub tool-definition event fixture and a model key for genuine run/status comparison are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Affected client/server versions, GitHub tool-definition event fixture and a model key for genuine run/status comparison are unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4670,+ "title": "WebSocketCallbackClient.stop() does not close a silent WebSocket and leaves its worker thread running",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4670",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Python WebSocketCallbackClient thread shutdown; not the Canvas browser WebSocket path.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4668,+ "title": "[Bug]: Planning file editor drops inherited observation and diff data",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4668",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A live planner create/edit yielding PlanningFileEditorObservation with old/new contents for the chat diff.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A live planner create/edit yielding PlanningFileEditorObservation with old/new contents for the chat diff.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4667,+ "title": "[Bug]: Tom processing history can checkpoint events that were never indexed",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4667",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Tom conversation index/checkpoint implementation; no mapped Canvas feature or supported Canvas agent path identified.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4666,+ "title": "[Bug]: Duplicate browser recording starts can orphan an active flush task",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4666",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "reason": "The report concerns the backend browser-use rrweb recording controller/flush-task lifecycle. Canvas has no backend recording/replay surface or duplicate-start action; control-openhands browser record is a separate verification-browser implementation and is not this SDK consumer.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4665,+ "title": "[Bug]: Skills marketplace returns empty against the current extensions manifest layout",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4665",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "historical-only",+ "reason": "Canvas SkillsService.getSkills uses the bundled @openhands/extensions catalog and the local /api/skills list with load_public:false. It does not call the SDK skills marketplace loader/endpoint whose default.json assumption is reported.",+ "family": null,+ "map_file": null,+ "issue_state": "open",+ "previous_state": "out-of-scope",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4664,+ "title": "[Bug]: OpenHandsCloudWorkspace resume leaves the active client on stale connection data",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4664",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "OpenHandsCloudWorkspace is a Python workspace object; Canvas uses TypeScript clients and Cloud runtime API transport, not that object or its cached Python connection fields.",+ "family": null,+ "map_file": null,+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "scope-reconfirmed",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4663,+ "title": "[Bug]: Python glob fallback mutates the process-wide working directory",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4663",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "Concurrent real glob/tool calls in distinct workspaces with the Python fallback active.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Concurrent real glob/tool calls in distinct workspaces with the Python fallback active.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4661,+ "title": "[Bug]: Process-global subagent registry leaks definitions across conversations",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4661",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "Two native task-enabled conversations with same-name project subagent definitions and a working model for delegation.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Two native task-enabled conversations with same-name project subagent definitions and a working model for delegation.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4660,+ "title": "[Bug]: RouterLLM bypasses model selection for async completions",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4660",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "basis": "historical-only",+ "reason": "Canvas Model Router uses meta-profiles and buildRouterAtStartSystemSuffix to guide profile selection; it does not instantiate the Python RouterLLM or call its completion/acompletion methods. The reported SDK class path is not its implementation.",+ "family": null,+ "map_file": null,+ "issue_state": "open",+ "previous_state": "out-of-scope",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4659,+ "title": "[Bug]: Concurrent stdout and stderr chunks can reuse BashOutput order values",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4659",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "reason": "Canvas BashService.listOutputs reads every result page sorted by timestamp, and no Canvas consumer polls BashOutput with order__gt. The reported duplicate order values affect reconstruction clients using that cursor; the current Canvas logs consumer does not use the affected order contract.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4658,+ "title": "[Bug]: Deferred initialization failure leaves a partially initialized runtime",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4658",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "feature_ids": [],+ "reason": "The report is transactionality/resource unwind in POST /api/init for a deferred warm-pool coordinator. Canvas neither calls that init endpoint nor owns its staged runtime publication, and no Canvas-observable failure is given in the report; this is a coordinator/Agent Server lifecycle contract before the UI attaches.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4629,+ "title": "[Bug]: Retired gemini-cli OAuth outranks a working API key, and the ACP Gemini pin is badly stale",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4629",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F13",+ "feature_ids": [+ "F13.acp-credentials"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current ACP dynamic model discovery added a new consumer, but the Gemini empty authfile diagnostic is distinct from successful login/model discovery. No fresh banner check was completed."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "line": 85+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4613,+ "title": "[Bug]: OpenAI provider prefix stripped twice for namespaced model IDs in Responses API",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4613",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.custom-model"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 174+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4583,+ "title": "[Bug]: file editor insert concatenates onto last line when file has no trailing newline",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4583",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working model invoking file-editor insert at EOF on a file with no trailing newline.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model invoking file-editor insert at EOF on a file with no trailing newline.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4575,+ "title": "[Bug]: Agent terminal commands inherit server process priority",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4575",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "basis": "retained-historical-note",+ "reason": "A model key and controlled bounded workload/resource envelope for a real Canvas terminal run are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "A model key and controlled bounded workload/resource envelope for a real Canvas terminal run are unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4562,+ "title": "[Bug]: workspace files return 401 on private-LAN HTTP because the session cookie is rejected",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4562",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F08",+ "feature_ids": [+ "F08.files-load-error"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "basis": "known-issue",+ "reason": "A plain-HTTP non-loopback same-origin browser deployment with real workspace-cookie observation.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A plain-HTTP non-loopback same-origin browser deployment with real workspace-cookie observation.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4558,+ "title": "Encrypted LLM profiles are decrypted only on the conversation path \u2014 FallbackStrategy (and sub-agents) load them without the cipher",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4558",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.error-events"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working encrypted saved fallback profile and real primary-model failure that triggers fallback resolution.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working encrypted saved fallback profile and real primary-model failure that triggers fallback resolution.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4540,+ "title": "[Bug]: Tool calls return incorrect/weird syntax with Synthetic Provider",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4540",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "Synthetic provider credentials and a GLM/Kimi turn emitting the raw tool_call wrapper.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "Synthetic provider credentials and a GLM/Kimi turn emitting the raw tool_call wrapper.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4537,+ "title": "[Bug]: TaskToolSet delegation holds the parent ConversationState lock for the whole sub-agent run, saturating the executor pool and freezing the conversation list on agent canvas UI",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4537",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "known-issue",+ "reason": "A working model with native TaskTool delegation holding the parent state lock during list reload.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working model with native TaskTool delegation holding the parent state lock during list reload.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4533,+ "title": "Conversation launch drops LLM api_key when the seeded `default` agent profile is active (litellm AuthenticationError)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4533",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F13",+ "feature_ids": [+ "F13.active-drives-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "basis": "live-not-reproduced",+ "reason": "Four actual seeded-default UI launches retained the dummy LLM key in runtime state and reached only refused-loopback model errors. This verifies propagation, not successful provider authentication.",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4533#issuecomment-6121790768",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-reproduced",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @/private/tmp/odie-18309-QA_LOOPBACK.json",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands api GET /api/agent-profiles/default --pick profile.llm_profile_ref",+ "control-openhands api GET /api/settings --pick llm_api_key_is_set",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation status fda5a4cd-9c39-4091-8b9a-5512f4c48642",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation status 950e5e83-f56d-4dbb-9aab-b2ccc78f1972",+ "python3 - <<'PY_KEY'\nimport json, os\nfrom pathlib import Path\nfor p in (Path(os.environ['OH_VERIFY_RUN']) / 'private').rglob('base_state.json'):\n d=json.loads(p.read_text())\n llm=d.get('agent',{}).get('llm',{})\n if llm.get('model')=='openai/gpt-4o': print({'conversation':p.parent.name,'model':llm.get('model'),'key_present':bool(llm.get('api_key'))})\nPY_KEY",+ "control-openhands api DELETE /api/profiles/QA_LOOPBACK --write",+ "control-openhands doctor"+ ],+ "evidence": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker/evidence/F13.active-drives-conversation/default-key-diagnostic.json"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Two seeded-default UI launches preserved the dummy key in runtime state and reached loopback connection refusal. This verifies key propagation, not successful provider authentication.",+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "/private/tmp/odie-18309-QA_LOOPBACK.json",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "api_key": "<dummy value read from private dummy-key file>",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ }+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "SDK1.55/#5983 now requires/restores an active AgentProfile; old default-inline key observation must be rerun on currentpin (root owns fresh replay)."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Arrange QA_ROOT_BASE dummy profile with base_url127.0.0.1:1/v1; activate the seeded default OpenHands agent",+ "control-openhands fixture skill --name qa-root-legacy --trigger qa-root-legacy-ping",+ "Move only this run-owned skill into <run>/private/skills",+ "control-openhands browser goto /skills",+ "control-openhands conversation start --prompt \"qa-root-legacy-ping. Reply with exactly pong. Do not run any tools.\"",+ "For datetime case: PATCH agent_settings_diff.agent_context.current_datetime=2001-01-01; repeat two UI starts; restore original",+ "Read owned base_state/event files; emit only key-present, datetime and skill-match diagnostics"+ ],+ "observations": {+ "initializations": [+ {+ "conversation": "2c860cb9-1352-4a51-957f-20641757b938",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T04:24:42.885562Z",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "9ee3edc4-d55c-42fc-b778-2b288c68686b",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T04:24:44.293887Z",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "546e89fc-f933-46d7-9d27-c6d1a3c946c9",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T06:23:21.923984+02:00",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "415433ba-f805-4da2-bebd-f6fa43f6e081",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T06:23:20.216502+02:00",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ }+ ]+ },+ "artifacts": [],+ "no_successful_model_inference": true+ },+ "current_result": "pass",+ "fresh_live_outcome": "not-reproduced",+ "comment_id": 6121790768,+ "refresh_published_at": "2026-10-16T04:40:24Z",+ "body_sha256": "4e17b0ef2232e3ad3dd28f3ac414b8ee35d107a4fab3f5312283dc26f24e7f7e",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "Arrange QA_ROOT_BASE dummy profile with base_url127.0.0.1:1/v1; activate the seeded default OpenHands agent",+ "control-openhands fixture skill --name qa-root-legacy --trigger qa-root-legacy-ping",+ "Move only this run-owned skill into <run>/private/skills",+ "control-openhands browser goto /skills",+ "control-openhands conversation start --prompt \"qa-root-legacy-ping. Reply with exactly pong. Do not run any tools.\"",+ "For datetime case: PATCH agent_settings_diff.agent_context.current_datetime=2001-01-01; repeat two UI starts; restore original",+ "Read owned base_state/event files; emit only key-present, datetime and skill-match diagnostics"+ ],+ "artifacts": [],+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4532,+ "title": "[Bug]: Reopening a confirmation-paused conversation drops the assistant tool_use message from the next request",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4532",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.confirmation-mode"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A real confirmation-paused action persisted across backend restart, then approval and a working provider call.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real confirmation-paused action persisted across backend restart, then approval and a working provider call.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4511,+ "title": "[Bug]: `Message.to_chat_dict` emits `cache_control` when caching is disabled",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4511",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.schema-validation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 175+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4498,+ "title": "Incremental view path skips enforce_properties after condensation, leaving orphaned tool_results/tool_uses for the LLM",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4498",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "A real condensation splitting an action/result pair, followed by the next provider request.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A real condensation splitting an action/result pair, followed by the next provider request.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4388,+ "title": "agent-server: search_bash_events pagination wrong when order__gt filter is active \u2014 pages undersized, clients loop on empty next-pages",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4388",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": true,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": "F23",+ "feature_ids": [+ "F23.run-status-polling",+ "F23.run-logs-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "known-issue",+ "reason": "Frontend BashService does not send order__gt, but automation 1.20 launch preparation and prompt/plugin preset clones use (Async)RemoteWorkspace.execute_command. Its remote_workspace_mixin output generator sends order__gt after received chunks. A bounded genuine worker/preparation or Git clone crossing the 100-event output boundary must be correlated with final BashOutput and Canvas run status; frontend Logs-only paging and a one-output command do not cover it.",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Frontend BashService does not send order__gt, but automation 1.20 launch preparation and prompt/plugin preset clones use (Async)RemoteWorkspace.execute_command. Its remote_workspace_mixin output generator sends order__gt after received chunks. A bounded genuine worker/preparation or Git clone crossing the 100-event output boundary must be correlated with final BashOutput and Canvas run status; frontend Logs-only paging and a one-output command do not cover it.",+ "verification_detail": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4382,+ "title": "[Bug]: ACP derived cost ignores cache/thought buckets, silently zeroes unknown models, and can inflate cost when UsageUpdate cost stalls",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4382",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F05",+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "basis": "known-issue",+ "reason": "An authenticated ACP provider emitting cache/thought token usage or reset/stalled cumulative cost updates.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "An authenticated ACP provider emitting cache/thought token usage or reset/stalled cumulative cost updates.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4363,+ "title": "InstallationManager.update() silently drops pinned refs (should reconcile, not force ref=None)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4363",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F19",+ "feature_ids": [+ "F19.refresh"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "live-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-plugins.tsx",+ "src/routes/launch.tsx",+ "src/components/features/plugins/",+ "src/components/features/launch/",+ "src/utils/plugin-launch-url.ts",+ "src/hooks/mutation/use-install-plugin.ts",+ "src/hooks/mutation/use-set-plugin-enabled.ts",+ "src/hooks/mutation/use-uninstall-plugin.ts",+ "src/hooks/mutation/use-refresh-plugin.ts"+ ],+ "changed_canvas_source_paths": [],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F19-plugins.md",+ "line": 96+ }+ ],+ "map_note_present": true,+ "verification_status": "historical-reproduction"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4260,+ "title": "[Bug]: Cloud Automation PR reviewer intermittently posts no review \u2014 SecretStr webhook serialization error (non-blocking) + 600s run ceiling killing stalled agent loop (blocker)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4260",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F23",+ "feature_ids": [+ "F23.run-linked-conversation",+ "F23.run-logs-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md",+ "basis": "retained-historical-note",+ "reason": "Owned Cloud reviewer/webhook deployment, authorized disposable review target and model credentials are unavailable.",+ "issue_state": "open",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "current_result": "not-run",+ "fresh_reproduction_claim": false,+ "disposition": "dated-note-retained-no-fresh-runtime-claim",+ "verification_status": "blocked",+ "verification_detail": "historical-only-no-current-replay",+ "prerequisite": "Owned Cloud reviewer/webhook deployment, authorized disposable review target and model credentials are unavailable."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4252,+ "title": "[Bug]: New added Global Skills dont get loaded into OpenHands (CLI install) and not available in WebUI",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4252",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F18",+ "feature_ids": [+ "F18.personal-skill-dirs"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "basis": "live-not-reproduced",+ "reason": "A persistence-folder skill appeared in the real Skills page and in initialized conversation events on four UI launches; no successful model reply or historical Docker mount behavior is claimed.",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4252#issuecomment-6121791082",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-reproduced",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @/private/tmp/odie-18309-QA_LOOPBACK.json",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands fixture skill --name qa-legacy --trigger qa-legacy-ping",+ "mkdir -p \"$OH_VERIFY_RUN/private/skills\"",+ "mv \"$OH_VERIFY_RUN/private/home/.agents/skills/qa-legacy\" \"$OH_VERIFY_RUN/private/skills/qa-legacy\"",+ "control-openhands browser goto /skills",+ "control-openhands browser reload",+ "control-openhands browser count 'testid=skill-card-qa-legacy'",+ "control-openhands browser attr 'testid=skill-source-qa-legacy' title",+ "control-openhands conversation start --prompt \"qa-legacy-ping. Do not run tools; reply in one line.\" --timeout 60",+ "control-openhands conversation events 099f2448-1df7-4f7a-a702-7248dcf8a961 --grep qa-legacy --from-start --last 4",+ "control-openhands conversation start --prompt \"qa-legacy-ping. Do not run tools; reply in one line.\" --timeout 60",+ "control-openhands conversation events ca80b3a6-7882-4821-b591-a09c15b16210 --grep qa-legacy --from-start --last 4",+ "control-openhands browser screenshot --feature F18.personal-skill-dirs --name 4252-legacy-skill-loaded",+ "python3 - <<'PY_CLEAN'\nimport os,shutil\nfrom pathlib import Path\np=Path(os.environ['OH_VERIFY_RUN'])/'private/skills/qa-legacy'\nif p.exists():shutil.rmtree(p)\nPY_CLEAN",+ "control-openhands doctor"+ ],+ "evidence": [+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker/evidence/F18.personal-skill-dirs/4252-legacy-skill-loaded.png"+ ],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "A persistence-folder skill appeared in the UI and in available/activated skills in two actual initialized chats. This verifies discovery before the model reply, not historical Docker mounts.",+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "/private/tmp/odie-18309-QA_LOOPBACK.json",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "api_key": "<dummy value read from private dummy-key file>",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ }+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-settings.tsx",+ "src/components/features/skills/",+ "src/constants/skills-docs.ts",+ "src/hooks/use-skill-enablement.ts",+ "src/utils/skill-enablement.ts",+ "src/hooks/use-launch-skill-in-chat.ts",+ "src/api/skills-service.ts",+ "src/components/features/chat/skill-install-restart-banner.tsx",+ "src/utils/skill-install-events.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/skills/skill-type-badge.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "SDK1.55/#5981 changes profile public-skill resolution and default seeding preserves disabledskills via#5744; historical persistence-folder activation needs fresh currentlaunch (rootreplay)."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Arrange QA_ROOT_BASE dummy profile with base_url127.0.0.1:1/v1; activate the seeded default OpenHands agent",+ "control-openhands fixture skill --name qa-root-legacy --trigger qa-root-legacy-ping",+ "Move only this run-owned skill into <run>/private/skills",+ "control-openhands browser goto /skills",+ "control-openhands conversation start --prompt \"qa-root-legacy-ping. Reply with exactly pong. Do not run any tools.\"",+ "For datetime case: PATCH agent_settings_diff.agent_context.current_datetime=2001-01-01; repeat two UI starts; restore original",+ "Read owned base_state/event files; emit only key-present, datetime and skill-match diagnostics"+ ],+ "observations": {+ "initializations": [+ {+ "conversation": "2c860cb9-1352-4a51-957f-20641757b938",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T04:24:42.885562Z",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "9ee3edc4-d55c-42fc-b778-2b288c68686b",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T04:24:44.293887Z",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "546e89fc-f933-46d7-9d27-c6d1a3c946c9",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T06:23:21.923984+02:00",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ },+ {+ "conversation": "415433ba-f805-4da2-bebd-f6fa43f6e081",+ "model": "openai/gpt-4o",+ "api_key_present": true,+ "runtime_datetime": "2026-10-16T06:23:20.216502+02:00",+ "skill_in_events": true,+ "old_date_in_events": false,+ "execution_status": "error"+ }+ ]+ },+ "artifacts": [],+ "no_successful_model_inference": true+ },+ "current_result": "pass",+ "fresh_live_outcome": "not-reproduced",+ "comment_id": 6121791082,+ "refresh_published_at": "2026-10-16T04:40:25Z",+ "body_sha256": "21584678fb205b869949eadaf7c77fd318618a3843cc02daefadf4489b5e541c",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "Arrange QA_ROOT_BASE dummy profile with base_url127.0.0.1:1/v1; activate the seeded default OpenHands agent",+ "control-openhands fixture skill --name qa-root-legacy --trigger qa-root-legacy-ping",+ "Move only this run-owned skill into <run>/private/skills",+ "control-openhands browser goto /skills",+ "control-openhands conversation start --prompt \"qa-root-legacy-ping. Reply with exactly pong. Do not run any tools.\"",+ "For datetime case: PATCH agent_settings_diff.agent_context.current_datetime=2001-01-01; repeat two UI starts; restore original",+ "Read owned base_state/event files; emit only key-present, datetime and skill-match diagnostics"+ ],+ "artifacts": [],+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4250,+ "title": "[Bug]: Workers AI broken",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4250",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.custom-model",+ "F10.default-used"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 176+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4247,+ "title": "[Bug]: LLM Provider NOT provided with LM Studio",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4247",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F10",+ "feature_ids": [+ "F10.custom-model"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "basis": "historical-only",+ "reason": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "issue_state": "open",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-run",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Current issue remains open. No targeted shipped fix identified in the Canvas owner-path/SDKv1.54\u2192v1.55 release review; preserve dated note and stated OS/account/endpoint limits without asserting fresh failure."+ },+ "current_map_refs": [+ {+ "path": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "line": 177+ }+ ],+ "map_note_present": true,+ "prerequisite": "Preserve the Oct12 observation and its exact limits as history; SDK pin changed1.54\u21921.55, so this row is not a current reproduction or current not-reproduced verdict.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4246,+ "title": "[Bug]:bserved Behavior: The agent remains idle. No visual feedback or error notification is shown in the interface.",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4246",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "not-reproduced",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "not-reproduced"+ },+ "family": "F17",+ "feature_ids": [+ "F17.agent-uses-server"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "basis": "live-not-reproduced",+ "reason": "Two actual90s stdio hang fixtures left initialization after the real MCP timeout, persisted system/user events and reached visible loopback model error. This settles the current Local initialization loop only; successful replies and historical Cloud/SSE remain unverified.",+ "comment_url": "https://github.com/OpenHands/software-agent-sdk/issues/4246#issuecomment-6121791434",+ "issue_state": "open",+ "previous_state": "not-reproduced",+ "previous_basis": "live-not-reproduced",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_verdict": "not-reproduced",+ "needs_current_replay": false,+ "historical_replay_reference": "/private/tmp/odie-18309-settings-results.json",+ "historical_live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-settings-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "backend": "localAgentServer1.54.0",+ "viewport": "1440x1000",+ "passes": 2,+ "steps": [+ "control-openhands api POST /api/profiles/QA_LOOPBACK --write --data @/private/tmp/odie-18309-QA_LOOPBACK.json",+ "control-openhands api POST /api/profiles/QA_LOOPBACK/activate --write",+ "control-openhands fixture mcp-server --name qa-settings-echo",+ "control-openhands api POST /api/settings/mcp/qa_settings_echo --write --data @<echo-config.json>",+ "python3 - <<'PY_SLOW'\nimport os\nfrom pathlib import Path\np=Path(os.environ['OH_VERIFY_RUN'])/'workspace/qa-settings-slow/server.mjs'\np.parent.mkdir(parents=True,exist_ok=True)\np.write_text('process.stdin.resume();setTimeout(()=>process.exit(0),90000);\\n')\nPY_SLOW",+ "control-openhands api POST /api/settings/mcp/qa_slow --write --data @/private/tmp/odie-18309-mcp-slow.json",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation status c08b83fc-1753-4b70-8f6a-59396e9205ce",+ "control-openhands conversation events c08b83fc-1753-4b70-8f6a-59396e9205ce --kinds SystemPromptEvent,MessageEvent --from-start --last 3",+ "control-openhands conversation start --prompt \"Reply with exactly pong.\" --timeout 60",+ "control-openhands conversation status da523c3d-3dd1-4ce9-b592-90ba99477aae",+ "control-openhands conversation events da523c3d-3dd1-4ce9-b592-90ba99477aae --kinds SystemPromptEvent,MessageEvent --from-start --last 3",+ "control-openhands api DELETE /api/settings/mcp/qa_slow --write",+ "control-openhands api DELETE /api/settings/mcp/qa_settings_echo --write",+ "control-openhands doctor"+ ],+ "evidence": [],+ "teardown": {+ "doctor_after_restore": "passed",+ "run_stopped": true,+ "ports_closed": [+ 18800,+ 18801,+ 18802,+ 18803,+ 19801+ ],+ "fixture_cleanup": "own QA profiles/connections/MCP/secrets/skills/auth/alias/plugin and cipher copies removed; seeded automation secret retained"+ },+ "observation": "Two actual stdio timeouts left initialization after about 30 seconds, persisted the prompt/user message and reached loopback model error. This settles the current Local initialization loop only, not successful replies or historical Cloud/SSE behavior; sibling tools were lost.",+ "replay_setup": {+ "working_directory": "/Users/enyst/.codex/worktrees/feature-map-backfill/odie",+ "path": "export PATH=\"$PWD/.agents/skills/verify-openhands/scripts:$PATH\"",+ "run": "export OH_VERIFY_RUN=<fresh owned launch directory; observed run is live.run>",+ "health": "control-openhands doctor",+ "onboarding": "control-openhands onboard --skip",+ "cleanup": "control-openhands stop --run \"$OH_VERIFY_RUN\"",+ "substitution": "Replace every historical conversation id in the readback commands with the actual id returned by its preceding conversation start. MCP config args must use the fresh run path and Node executable returned by its fixture command. Historical literal ids identify the captured observations.",+ "loopback_profile": {+ "file": "/private/tmp/odie-18309-QA_LOOPBACK.json",+ "body": {+ "llm": {+ "model": "openai/gpt-4o",+ "api_key": "<dummy value read from private dummy-key file>",+ "base_url": "http://127.0.0.1:1/v1",+ "num_retries": 0,+ "temperature": 0.4+ },+ "include_secrets": true+ },+ "note": "Arrange through body file; no real credentials or successful model replies. Restore previous active profile before deleting an active fixture."+ },+ "echo_config": {+ "file": "<echo-config.json>",+ "body": {+ "transport": "stdio",+ "command": "<Node executable returned by fixture mcp-server>",+ "args": [+ "<server.mjs returned by fixture mcp-server>"+ ],+ "enabled": true+ }+ },+ "slow_config": {+ "file": "/private/tmp/odie-18309-mcp-slow.json",+ "body": {+ "transport": "stdio",+ "command": "<Node executable returned by fixture mcp-server>",+ "args": [+ "<OH_VERIFY_RUN>/workspace/qa-settings-slow/server.mjs"+ ],+ "enabled": true+ }+ }+ }+ },+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "SDK1.55/#4838 changes MCP HTTP recovery but does not by itself prove the real90second hangingstdio initialization timeout; root currentreplay required."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "live": {+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "pins": {+ "agentServer": "1.55.0",+ "sdk": "1.55.0",+ "automation": "1.20.0"+ },+ "steps": [+ "Arrange bounded90s own stdio process plus healthy qa_echo",+ "Twice start a confined UI conversation; wait up to45s for error; inspect initial events",+ "Delete own slow/healthy MCP settings"+ ],+ "observations": {+ "results": [+ {+ "id": "585b0acf-6fbc-4c29-95ae-4fc1258d6b65",+ "status": "error"+ },+ {+ "id": "8a4315e6-b1f9-4dcc-b42f-4f70a24eb499",+ "status": "error"+ }+ ],+ "successful_inference": false+ },+ "artifacts": [],+ "no_successful_model_inference": true+ },+ "current_result": "pass",+ "fresh_live_outcome": "not-reproduced",+ "comment_id": 6121791434,+ "refresh_published_at": "2026-10-16T04:40:26Z",+ "body_sha256": "988a09b4495096cfbd0fe1062b18779df6bb2371c3bd0d00a6e8398f04922841",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "Arrange bounded90s own stdio process plus healthy qa_echo",+ "Twice start a confined UI conversation; wait up to45s for error; inspect initial events",+ "Delete own slow/healthy MCP settings"+ ],+ "artifacts": [],+ "fresh_reproduction_claim": false,+ "verification_status": "not-reproduced"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4245,+ "title": "[Bug]: Agent-Server Webhook Connection Failures Cause Container Crashes and Sandbox Connection Errors",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4245",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "out-of-scope"+ },+ "reason": "Legacy docker.openhands.dev/openhands/openhands v0 webhook deployment, absent from current Agent Canvas stack.",+ "basis": "retained-scope-review",+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4158,+ "title": "[Bug] switch_profile on an ACP conversation half-applies: state file updated but live session keeps old agent",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4158",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "out-of-scope",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "open",+ "family": null,+ "feature_ids": [],+ "map_file": null,+ "basis": "current-source",+ "reason": "Current ACP composer offers ACP model choices and not an OpenHands LLM-profile switcher; it cannot request the cross-agent-kind switch_profile described by the old1.4.0 report. Local OpenHands profile switching now uses fresh switchLLM configs, while ACP remains a separate picker.",+ "source_refs": [+ "src/hooks/use-chat-input-model-state.ts",+ "src/components/features/chat/components/chat-input-actions.tsx:234",+ "src/api/conversation-service/agent-server-conversation-service.api.ts:1201",+ "src/components/features/chat/components/chat-input-model.tsx:84"+ ],+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_map_refs": [],+ "map_note_present": false,+ "verification_status": "out-of-scope"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4080,+ "title": "One unregistered event kind fails the entire conversation load (should degrade per-event)",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4080",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F04",+ "feature_ids": [+ "F04.card"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md",+ "basis": "known-issue",+ "reason": "A copied persisted conversation with its original unregistered observation kind and missing tool-module registration.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A copied persisted conversation with its original unregistered observation kind and missing tool-module registration.",+ "verification_status": "blocked"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4077,+ "title": "Streaming: correctness & resource-safety bugs in the token/delta pipeline",+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4077",+ "issue_state_at_snapshot": "open",+ "ready_for_dev_at_snapshot": false,+ "state": "mapped",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "historical_record": {+ "file": "2026-10-12-open-bugs.json",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "outcome": "mapped"+ },+ "family": "F06",+ "feature_ids": [+ "F06.reload-mid-run"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "basis": "known-issue",+ "reason": "A working stream interrupted/retried or delivered to a controlled stalled subscriber, with delta/state-order capture.",+ "reviewed_at": "2026-10-16",+ "issue_state": "open",+ "fresh_live_outcome": null,+ "disposition": "retained-source-consumer",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "prerequisite": "A working stream interrupted/retried or delivered to a controlled stalled subscriber, with delta/state-order capture.",+ "verification_status": "blocked"+ }+ ],+ "closed_issue_reconciliation": [+ {+ "repository": "OpenHands/OpenHands",+ "number": 18303,+ "url": "https://github.com/OpenHands/OpenHands/issues/18303",+ "title": "[Bug]: \"Subscription models are not loaded yet.\" is shown untranslated",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T13:46:18Z",+ "family": "F10",+ "feature_ids": [+ "F10.subscription"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "issue_state": "closed",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "outcome": "mapped",+ "basis": "historical-only",+ "current_verdict": "not-run",+ "reason": "Merged main#18308/76eadf61f6 uses SETTINGS$SUBSCRIPTION_MODELS_NOT_LOADED in settings and onboarding; upstream map explicitly source-checked2026-10-13. Preserve accepted behavior without a new slow-request UI claim.",+ "disposition": "inherited-main-accepted",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Currentmain#18308 translates both callsites; inherited accepted sourcecheck is datedOct13. Slowmodel-list branch not driven today."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "current_map_disposition": "Resolved issue history removed; normal current expectations retained. Acceptance/source evidence remains in this dated report."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17955,+ "url": "https://github.com/OpenHands/OpenHands/issues/17955",+ "title": "[Bug]: Add LLM Profile shows the agent's current LLM options (such as temperature) but saves the new profile without them",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-14T04:15:52Z",+ "family": "F10",+ "feature_ids": [+ "F10.create-prefill"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md",+ "issue_state": "closed",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "outcome": "mapped",+ "basis": "historical-only",+ "current_verdict": "not-run",+ "reason": "Merged main#18422/bf483c12ef persists displayed new-form values. Fresh real Chrome linked create saved inherited temperature0.4, confirmed GET. Baseline/fix paired known-bug proof was not run, so this is current acceptance, not a new fixed verdict.",+ "disposition": "live-current-acceptance",+ "source_review": {+ "canvas_sources": [+ "src/routes/llm-settings.tsx",+ "src/components/features/settings/llm-profiles/",+ "src/components/shared/modals/settings/model-selector.tsx",+ "src/components/features/settings/settings-input.tsx",+ "src/components/features/settings/llm-settings/openai-subscription-auth-card.tsx",+ "src/components/features/settings/sdk-settings/",+ "src/constants/llm-provider-connection.ts",+ "src/api/profiles-service/profiles-service.api.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/llm-profiles/llm-profile-form.ts",+ "src/components/features/settings/llm-profiles/llm-settings-local-view.tsx",+ "src/components/features/settings/llm-profiles/profiles-body.tsx",+ "src/components/features/settings/llm-profiles/provider-connections-manager.tsx",+ "src/routes/llm-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Currentmain#18422 accepted expectations plus fresh provider-linked Save persisted displayedtemperature0.4. This is current acceptance only, no paired baseline/fix proof."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "current_map_disposition": "Resolved issue history removed; normal current expectations retained. Acceptance/source evidence remains in this dated report."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17920,+ "url": "https://github.com/OpenHands/OpenHands/issues/17920",+ "title": "[Bug]: A rejected (Cancel) action leaves its event group spinning as \"k/N actions completed\" after the conversation is idle",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T04:59:30Z",+ "state": "mapped",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.confirmation-mode"+ ],+ "basis": "accepted-main-expectation",+ "reason": "Merged17995/current-main maintenance preserves resolved/rejected groups and normal expectations. Frozen October12 report retains history; no fresh runtime proof claimed.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "retired-by-current-main",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "src/utils/handle-event-for-ui.ts"+ ]+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17900,+ "url": "https://github.com/OpenHands/OpenHands/issues/17900",+ "title": "[Bug]: Settings switches are skipped by Tab and invisible to screen readers (SettingsSwitch renders <input hidden>)",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-14T19:30:13Z",+ "family": "F16",+ "feature_ids": [+ "F16.keyboard"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F16-application-settings.md",+ "issue_state": "closed",+ "previous_state": "mapped",+ "previous_basis": "existing-note",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "outcome": "mapped",+ "basis": "historical-only",+ "current_verdict": "not-run",+ "reason": "On the actual settings page in the existing Chrome session, Escape from Color Theme followed by Tab focused the Analytics checkbox; the next Tab focused Sound and Space toggled it on and off. The settings form returned to its unchanged state. This is a current acceptance observation, not a baseline/fix proof.",+ "disposition": "current-keyboard-scenario-pass",+ "source_review": {+ "canvas_sources": [+ "src/routes/app-settings.tsx",+ "src/components/features/settings/app-settings/",+ "src/components/features/settings/settings-switch.tsx",+ "src/themes/color-theme/",+ "src/utils/transcription-endpoint-storage.ts",+ "src/api/settings-service/settings-service.api.ts",+ "src/services/telemetry.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/settings-switch.tsx",+ "src/routes/app-settings.tsx",+ "src/services/telemetry.ts",+ "src/themes/color-theme/definitions/light-plus.ts",+ "src/themes/color-theme/definitions/solarized-light.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Currentmain#18370 switches are native sr-only checkboxes with visible keyboardfocus label; accepted main behavior retained, fresh keyboardreplay incomplete."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "current_map_disposition": "Resolved issue history removed; normal current expectations retained. Acceptance/source evidence remains in this dated report.",+ "current_result": "pass",+ "checked_at": "2026-10-16",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-settings-root",+ "steps": [+ "Open /settings/app in the existing Chrome session",+ "Open Color Theme, press Escape, then Tab: Analytics checkbox receives focus",+ "Press Tab: Sound checkbox receives focus; Space toggles on, Space toggles off",+ "Confirm Save Changes is disabled again"+ ],+ "artifacts": [+ "/private/tmp/odie-18334-refresh/oct16-keyboard-settings.png"+ ]+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17485,+ "url": "https://github.com/OpenHands/OpenHands/issues/17485",+ "title": "Git info probe on the shared bash-events socket can resolve with another client's command output",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:57:46Z",+ "state": "out-of-scope",+ "family": null,+ "map_file": null,+ "feature_ids": [],+ "basis": "source-scope",+ "reason": "The reported persistent shared bash-events correlation consumer was removed. Current useBashCommandRunner uses conversation-scoped BashClient.executeCommand.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "scope-reconfirmed",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17096,+ "url": "https://github.com/OpenHands/OpenHands/issues/17096",+ "title": "docs: DEVELOPMENT.md claims agent-server defaults to latest PyPI",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T15:01:14Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "Agent Server version default documentation drift; documentation only.",+ "references": [+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/17097",+ "title": "docs: correct agent-server default version in DEVELOPMENT.md",+ "state": "closed"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16748,+ "url": "https://github.com/OpenHands/OpenHands/issues/16748",+ "title": "[Bug]: CI red on main while every test passes \u2014 a late mock XHR throws ProgressEvent ReferenceError",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:59:18Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "MSW/jsdom teardown unhandled rejection; test runner lifecycle.",+ "references": [+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/16751",+ "title": "fix(ci): let a bug with no rendered surface prove itself with a command",+ "state": "closed"+ },+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/16776",+ "title": "fix(conversation-panel): render idle conversations as paused, not working (#16752)",+ "state": "closed"+ },+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/16805",+ "title": "fix: prevent late MSW requests during vitest teardown",+ "state": "closed"+ },+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/16778",+ "title": "fix: refresh commits after terminal commands",+ "state": "closed"+ },+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/16890",+ "title": "fix(test): drain MSW's in-flight requests instead of 30 event-loop turns",+ "state": "closed"+ },+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/16891",+ "title": "fix: scope plugin queries to the active backend",+ "state": "closed"+ },+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/16892",+ "title": "fix: scope workspace queries to the active backend",+ "state": "open"+ },+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/17023",+ "title": "fix: keep git changes fresh and unify workspace-files query keys (#16949)",+ "state": "closed"+ },+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/17033",+ "title": "fix: count unified diff hunk lines whose content starts with -- or ++ (#16979)",+ "state": "open"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16580,+ "url": "https://github.com/OpenHands/OpenHands/issues/16580",+ "title": "use-drag-resize: touchmove/touchend listeners registered with capture:true but removed without it \u2014 permanent listener accumulation on mobile after each drag",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T15:00:20Z",+ "state": "mapped",+ "family": "F05",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "feature_ids": [+ "F05.composer-resize"+ ],+ "basis": "historical-source-reconciliation",+ "reason": "Touch listener removal now uses capture=true, matching registration. Repeated actual touch drags remain unverified; mouse drag cannot prove this.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-source-changed-awaiting-replay",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "src/hooks/use-drag-resize.ts"+ ]+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16579,+ "url": "https://github.com/OpenHands/OpenHands/issues/16579",+ "title": "use-load-older-events: stale conversationId closure after await injects wrong conversation events and permanently disables load-older for new conversation",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:58:29Z",+ "state": "mapped",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.load-older-history"+ ],+ "basis": "historical-source-reconciliation",+ "reason": "Active-load markers and loadedConversationId now guard abandoned older-history responses. Delayed-response navigation race needs genuine replay.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-source-changed-awaiting-replay",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "src/hooks/use-load-older-events.ts"+ ]+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16429,+ "url": "https://github.com/OpenHands/OpenHands/issues/16429",+ "title": "fix: grant release-ready contents write permission",+ "issue_state_at_snapshot": "closed",+ "state_reason": "not_planned",+ "closed_at": "2026-10-12T13:34:49Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "Release-ready workflow permissions; repository CI tooling.",+ "references": [+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/16086",+ "title": "fix: grant release-ready contents write permission",+ "state": "closed"+ },+ {+ "url": "https://github.com/harshitethic/OpenHands/pull/3",+ "title": "fix(ci): grant release-ready contents write permission",+ "state": "open"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 15433,+ "url": "https://github.com/OpenHands/OpenHands/issues/15433",+ "title": "[Bug]: user messages can split streaming text",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:58:12Z",+ "state": "mapped",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.user-and-agent-messages"+ ],+ "basis": "historical-source-reconciliation",+ "reason": "Main replaced the streaming presentation path. Successful provider streaming plus an interleaved user message remains unverified.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-source-changed-awaiting-replay",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "src/hooks/use-streamed-text.ts"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5646,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5646",+ "title": "[Bug]: Canvas App backend stays 'unhealthy' after one slow health probe instead of returning to 'ready'",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-13T22:37:52Z",+ "family": "F20",+ "feature_ids": [+ "F20.app-backend-view"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "issue_state": "closed",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "reason": "Closed after SDK PR 5759; installed 1.55.0 source has re-probing for live unhealthy backends. No fresh baseline/fix proof. The Oct12 default stack lacked the app-backend bridge capability; a reviewed platform artifact and admitted bridge deployment are still required. SDK1.55.0 backend platform selection supports only Linux; this native macOS host cannot execute that fixture.",+ "historical_record": ".agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.json",+ "disposition": "closed-source-included-runtime-blocked-on-linux",+ "pin_status": "included-in-sdk-1.55.0-by-installed-source-review",+ "verification_status": "blocked",+ "prerequisite": "Closed after SDK PR 5759; installed 1.55.0 source has re-probing for live unhealthy backends. No fresh baseline/fix proof. The Oct12 default stack lacked the app-backend bridge capability; a reviewed platform artifact and admitted bridge deployment are still required. SDK1.55.0 backend platform selection supports only Linux; this native macOS host cannot execute that fixture.",+ "map_cleanup": "Resolved issue/history reference removed from the current feature map. Dated evidence remains in this backfill ledger; current expected behavior and distinct open issue prerequisites are retained."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5641,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5641",+ "title": "[Bug]: GET /api/git/changes returns [] for a non-repo directory holding a repository, not its changes",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-15T19:47:42Z",+ "state": "mapped",+ "family": "F08",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "basis": "live-reproduced",+ "reason": "On pinned1.55.0, actual Commits tab for verified plain parent workspace shows No commits yet; scoped /git/changes returned200 while owned child proj has modified README and untracked notes. Closed upstream fix is not in current pin.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18309-surface-worker",+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "evidence": [+ "evidence/F08.uncommitted-changes/sdk-5641-settled.png"+ ],+ "steps": [+ "Arrange qa-nested/proj Git repo with committed README.md, modify README.md and add notes.txt.",+ "workspace open <run>/workspace/qa-nested; create empty conversation through sidebar workspace launch.",+ "Server creates empty qa-nested/.git; remove only that disposable empty .git to restore the reported non-repo parent layout.",+ "browser reload; open right panel on Commits; browser wait-text \"No commits yet\".",+ "browser network /git/ => scoped changes200; real child repo remains dirty."+ ],+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": {+ "number": 5641,+ "result": "fail",+ "actual": "On pinned1.55.0, actual Commits tab for verified plain parent workspace shows No commits yet; scoped /git/changes returned200 while owned child proj has modified README and untracked notes. Closed upstream fix is not in current pin.",+ "expected": "Changes in the single nested repository remain discoverable.",+ "evidence": [+ "evidence/F08.uncommitted-changes/nested-parent-5641.png"+ ],+ "steps": [+ "Create owned qa-surface-nested/proj Git repo, commit README, then modify it and add notes.txt.",+ "Create idle dummy conversation in plain parent. Verify and remove only its auto-created empty .git (canonical path and zero tracked files).",+ "Reload, open right panel on saved Commits tab. UI No commits yet; /git/changes200; child git status still M README and ??notes."+ ],+ "at": "2026-10-16T04:44:32.149764+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "known_bug_contract": {+ "id": "F08.uncommitted-changes",+ "name": "nested-repository",+ "status": "xfail",+ "sha": "fd68e17f5c9b522e",+ "assertion": {+ "kind": "mismatch",+ "expected": "1",+ "actual": "0",+ "bugShows": "0",+ "bugChecks": [+ {+ "clause": "count eq \"0\"",+ "actual": "0",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F08.uncommitted-changes-nested-repository-1792126580813-1.json"+ }+ },+ "disposition": "closed-upstream-not-in-current-pin",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "pin": "SDK1.55.0",+ "verification": "fail",+ "known_bug_contract": {+ "id": "F08.uncommitted-changes",+ "name": "nested-repository",+ "status": "xfail",+ "sha": "fd68e17f5c9b522e",+ "assertion": {+ "kind": "mismatch",+ "expected": "1",+ "actual": "0",+ "bugShows": "0",+ "bugChecks": [+ {+ "clause": "count eq \"0\"",+ "actual": "0",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F08.uncommitted-changes-nested-repository-1792126580813-1.json"+ }+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5640,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5640",+ "title": "[Bug]: download-trajectory returns 404 after POST /api/init sets conversations_path (expected 200 zip)",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-15T23:13:25Z",+ "state": "mapped",+ "family": "F07",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md",+ "feature_ids": [+ "F07.download-zip"+ ],+ "basis": "pinned-source-review",+ "reason": "Closed upstream, but pinned1.55 still uses get_default_config().conversations_path for trajectory ZIP. Deferred-init custom storage remains supported version distinction.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-upstream-not-in-current-pin",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "pin": "SDK1.55.0"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5637,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5637",+ "title": "[Bug]: Canvas backend outlives a SIGKILLed agent-server; restart reports it stopped and starts a second one",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-14T18:59:19Z",+ "family": "F20",+ "feature_ids": [+ "F20.app-backend-view"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md",+ "issue_state": "closed",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "reason": "Closed after SDK PR 5774; installed 1.55.0 source has persisted backend process records and startup orphan reaping. No fresh baseline/fix proof. The Oct12 default stack lacked the app-backend bridge capability; a reviewed platform artifact and admitted bridge deployment are still required. SDK1.55.0 backend platform selection supports only Linux; this native macOS host cannot execute that fixture.",+ "historical_record": ".agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.json",+ "disposition": "closed-source-included-runtime-blocked-on-linux",+ "pin_status": "included-in-sdk-1.55.0-by-installed-source-review",+ "verification_status": "blocked",+ "prerequisite": "Closed after SDK PR 5774; installed 1.55.0 source has persisted backend process records and startup orphan reaping. No fresh baseline/fix proof. The Oct12 default stack lacked the app-backend bridge capability; a reviewed platform artifact and admitted bridge deployment are still required. SDK1.55.0 backend platform selection supports only Linux; this native macOS host cannot execute that fixture.",+ "map_cleanup": "Resolved issue/history reference removed from the current feature map. Dated evidence remains in this backfill ledger; current expected behavior and distinct open issue prerequisites are retained."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5630,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5630",+ "title": "[Bug]: Sub-agent file naming a missing skill makes the first message return 500 instead of being skipped",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-13T21:39:50Z",+ "ready_for_dev_at_snapshot": true,+ "versions": {+ "agent_canvas": "1.26.0",+ "agent_server": "1.54.0",+ "automation": "1.19.3"+ },+ "cleanup": "Only owned fixture frontends stopped; copied fault-injected build removed; inert automations deactivated; bounded worker exited; normal launcher restored and doctored before stop.",+ "state": "mapped",+ "family": "F03",+ "feature_ids": [+ "F03.launch-in-workspace",+ "F03.create-error-toast"+ ],+ "basis": "scoped-live-not-reproduced",+ "reason": "Original owned project subagent references missing skill. Actual Home Local Repo launch on1.55 returnedPOST/conversations201, fetched metadata200 and opened conversation route; former missing-skill500 did not occur. Later dummy-model error is separate and expected. Pinned _register_definition catches ValueError, warns and skips invalid agent.",+ "run_id": "backfill-18309-integrations-worker",+ "steps": [+ "Use a fresh owned run at the recorded frozen revision; doctor and complete onboarding before the UI drive.",+ "- **Missing sub-agent skill (`F03.launch-in-workspace`, `F03.create-error-toast`).** In the run-owned `<run>/workspace/qa-bad-skill/.agents/agents/qa-bad-skill.md`, arrange YAML frontmatter with `name: qa-bad-skill`, `description: Bad skill` and `skills: qa-f27-missing-skill`, followed by body `Body`; do not install that missing skill. Record the original active LLM profile and autotitle setting. Arrange a dummy `qa-offline` profile through `control-openhands api POST /api/profiles/qa-offline --write --data '@<dummy-profile-json>'`: `llm` contains model `openai/qa-offline`, a dummy API key, base URL `http://127.0.0.1:9`, `num_retries: 0`, `timeout: 2` and `usage_id: qa-offline`, with `include_secrets: true`; activate it through `api POST /api/profiles/qa-offline/activate --write --data '{}'`. Set `conversation_settings_diff.autotitle` to `false` through `api PATCH /api/settings --write --data '@<settings-json>'`, so no provider call is possible. From Home open `testid=open-workspace-button`, `testid=workspace-dropdown` and `testid=add-workspaces-button`; use `testid=folder-browser-up` twice, `testid=folder-browser-entry-workspace`, `testid=folder-browser-entry-qa-bad-skill`, `testid=folder-browser-use` and `testid=workspace-launch-button` to confirm that folder. Fill `testid=chat-input` with `QA missing skill creation check`, then `control-openhands browser click 'testid=home-chat-launcher >> testid=submit-button' --observe 'role=status' --observe-ms 3000`. Expected: an invalid sub-agent is skipped with a warning or reported by an actionable 4xx; it does not turn conversation creation into a 500. Known failure (reproduced 2026-10-12, [OpenHands/software-agent-sdk#5630](https://github.com/OpenHands/software-agent-sdk/issues/5630)): `Creating conversation` is replaced by `Skill 'qa-f27-missing-skill' not found but was given to agent 'qa-bad-skill'.`; the loading state ends, the prompt remains and the URL stays `/`. Genuine message requests with `run: false` and creation with an initial message also return 500 before inference. Verified in `backfill-18309-integrations-worker` on `fa1f724ab5737a17453ad6081129eb91c5a423c6`, SDK 1.54.0; evidence `F03.create-error-toast/missing-skill-home-5630-2.png`. Remove only this run-owned agent file and test conversations afterwards, restore the original autotitle setting and active profile, then delete the dummy profile."+ ],+ "artifacts": [+ "evidence/F03.create-error-toast/missing-skill-home-5630-2.png",+ "evidence/F03.create-error-toast/bad-skill-5630.aria.txt"+ ],+ "source_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F03-home.md",+ "issue_state": "closed",+ "disposition": "closed-fix-in-pin-live-creation-pass",+ "source_refs": [+ "SDK1.55.0: openhands/sdk/subagent/registry.py"+ ],+ "verification_limit": "Fresh credential-free Home creation passed on1.55; later refused-loopback model error is separate, not successful provider proof.",+ "fresh_live_outcome": {+ "number": 5630,+ "result": "pass",+ "actual": "Original owned project subagent references missing skill. Actual Home Local Repo launch on1.55 returnedPOST/conversations201, fetched metadata200 and opened conversation route; former missing-skill500 did not occur. Later dummy-model error is separate and expected. Pinned _register_definition catches ValueError, warns and skips invalid agent.",+ "expected": "Bad project subagent does not turn conversation creation into500.",+ "evidence": [+ "evidence/F03.create-error-toast/missing-skill-current-pin-5630.png"+ ],+ "steps": [+ "Create owned .agents/agents/qa-bad-skill.md with name/description and skills: qa-f27-missing-skill.",+ "Keep active explicit refused-loopback QA_SURFACE_REFRESH profile and autotitle off.",+ "Home workspace open owned folder, select Local Repo, fill QA_SURFACE bad-skill pre-inference check, click Home Send.",+ "Actual network create201 and metadata200, URL /conversations/<new-id>. No successful provider response required."+ ],+ "limit": "Verifies creation/pre-inference regression only, not successful model output.",+ "at": "2026-10-16T04:58:19.335138+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F03.launch-in-workspace",+ "F03.create-error-toast"+ ]+ },+ "verification": "pass"+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5581,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5581",+ "title": "[Bug]: Agent-server startup kills live terminal sessions of other conversations sharing the same `TMUX_TMPDIR`",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-15T10:32:31Z",+ "family": "F27",+ "feature_ids": [+ "F27.terminal-output"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md",+ "issue_state": "closed",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "reason": "SDK PR5589 closed the report; installed 1.55.0 source uses the cross-process socket creation lock and ownership-aware stale-socket reclamation. Needs a bounded owned shared-TMUX_TMPDIR command fixture for current runtime proof; no current reproduction or fixed verdict is claimed.",+ "historical_record": ".agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.json",+ "disposition": "closed-awaiting-live-proof",+ "pin_status": "included-in-sdk-1.55.0-by-installed-source-review",+ "verification_status": "blocked",+ "prerequisite": "SDK PR5589 closed the report; installed 1.55.0 source uses the cross-process socket creation lock and ownership-aware stale-socket reclamation. Needs a bounded owned shared-TMUX_TMPDIR command fixture for current runtime proof; no current reproduction or fixed verdict is claimed.",+ "map_cleanup": "Resolved issue/history reference removed from the current feature map. Dated evidence remains in this backfill ledger; current expected behavior and distinct open issue prerequisites are retained."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5519,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5519",+ "title": "[Bug]: A scoped profile can't delegate to general-purpose because it lacks task_tracker",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-13T12:29:09Z",+ "family": "F13",+ "feature_ids": [+ "F13.openhands-options",+ "F13.active-drives-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "issue_state": "closed",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "outcome": "mapped",+ "basis": "historical-only",+ "current_verdict": "not-run",+ "reason": "SDK1.55 includes c1dfd9c52/#5520, scoped subagents retain conversation-local task_tracker. Current Canvas also still adds task_tracker during scoped profile Save. Remove open Known issue wording, preserving Oct12 report history; no fresh model-driven delegation proof.",+ "disposition": "fix-in-current-pin",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay.",+ "conclusion": "Scoped-subagent local-tool exceptionc1dfd9c52/#5520 IS in pinnedSDKv1.55.0. Retire open Knownissue wording; no authenticated delegated turn run today."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "current_map_disposition": "Resolved issue history removed; normal current expectations retained. Acceptance/source evidence remains in this dated report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5481,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5481",+ "title": "[Bug]:",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:23:43Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "Empty bug template identifies no behavior, reproduction, version or mapped Canvas feature; needs a concrete report before Canvas relevance can be established.",+ "references": [],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5268,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5268",+ "title": "[Bug]: openhands.sdk.conversation.exceptions.ConversationRunError: 'PromptTokensDetailsWrapper' object has no attribute 'cache_creation_tokens'",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:23:29Z",+ "state": "mapped",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "historical-source-reconciliation",+ "reason": "Pinned normalize_usage guards cache_creation_tokens with model_fields_set and hasattr, defaulting absent/null values to0. Original provider /btw or model response needs replay; source/closure is not live proof.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-source-changed-awaiting-provider-replay",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "SDK1.55.0: openhands/sdk/llm/utils/telemetry.py"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5213,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5213",+ "title": "[Bug]: ask_agent 500: `AttributeError: 'PromptTokensDetailsWrapper' object has no attribute 'cache_creation_tokens'` in telemetry `_cache_buckets` with litellm >= 1.95.1",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:23:11Z",+ "state": "mapped",+ "family": "F05",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F05-composer.md",+ "feature_ids": [+ "F05.slash-btw"+ ],+ "basis": "historical-source-reconciliation",+ "reason": "Pinned normalize_usage guards cache_creation_tokens with model_fields_set and hasattr, defaulting absent/null values to0. Original provider /btw or model response needs replay; source/closure is not live proof.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-source-changed-awaiting-provider-replay",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "SDK1.55.0: openhands/sdk/llm/utils/telemetry.py"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5168,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5168",+ "title": "AttributeError in Telemetry._cache_buckets for providers without prompt caching",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:22:40Z",+ "state": "mapped",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "historical-source-reconciliation",+ "reason": "Pinned normalize_usage guards cache_creation_tokens with model_fields_set and hasattr, defaulting absent/null values to0. Original provider /btw or model response needs replay; source/closure is not live proof.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-source-changed-awaiting-provider-replay",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "SDK1.55.0: openhands/sdk/llm/utils/telemetry.py"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5099,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5099",+ "title": "1.48.0: _cache_buckets raises AttributeError on litellm PromptTokensDetailsWrapper",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:22:10Z",+ "state": "mapped",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "historical-source-reconciliation",+ "reason": "Pinned normalize_usage guards cache_creation_tokens with model_fields_set and hasattr, defaulting absent/null values to0. Original provider /btw or model response needs replay; source/closure is not live proof.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-source-changed-awaiting-provider-replay",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "SDK1.55.0: openhands/sdk/llm/utils/telemetry.py"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5072,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/5072",+ "title": "fix(ts-client): keep Babel override consistent with lockfile",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:06:05Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "TypeScript Babel override/lockfile consistency for builds; no Canvas runtime feature.",+ "references": [+ {+ "url": "https://github.com/OpenHands/software-agent-sdk/pull/5073",+ "title": "chore(ts-client): reconcile Babel fix after Vitest migration",+ "state": "closed"+ },+ {+ "url": "https://github.com/OpenHands/software-agent-sdk/pull/5075",+ "title": "test(ts-client): migrate from Jest to Vitest",+ "state": "closed"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4910,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4910",+ "title": "[Bug]: ACP subprocess leaf process orphaned on shutdown on POSIX systems",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:21:28Z",+ "family": "F26",+ "feature_ids": [+ "F26.runtime-services"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md",+ "issue_state": "closed",+ "previous_outcome": "mapped",+ "historical_basis": "known-issue",+ "historical_evidence_revision": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "current_result": "blocked",+ "fresh_reproduction_claim": false,+ "reason": "SDK4910 was closed without a merged fixing reference; companion SDK4901 remains open and maps the same ACP descendant teardown scenario. Retain its exact owned provider-process prerequisite; closure alone is not runtime proof.",+ "historical_record": ".agents/skills/verify-openhands/references/backfills/2026-10-12-open-bugs.json",+ "disposition": "closed-awaiting-live-proof",+ "pin_status": "closure-is-not-proof-of-a-merged-fix",+ "verification_status": "blocked",+ "prerequisite": "SDK4910 was closed without a merged fixing reference; companion SDK4901 remains open and maps the same ACP descendant teardown scenario. Retain its exact owned provider-process prerequisite; closure alone is not runtime proof.",+ "map_cleanup": "Resolved issue/history reference removed from the current feature map. Dated evidence remains in this backfill ledger; current expected behavior and distinct open issue prerequisites are retained."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4887,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4887",+ "title": "Centralize release publication dispatches",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:21:59Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "SDK release publication workflow dispatch; no Canvas feature.",+ "references": [+ {+ "url": "https://github.com/OpenHands/software-agent-sdk/pull/4886",+ "title": "fix(ci): centralize release publication dispatches",+ "state": "closed"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4798,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4798",+ "title": "TypeScript client endpoint audit fails on every fork PR: comment POST 403s with a read-only token",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:20:25Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "TypeScript client endpoint-audit CI comment permissions; no Canvas feature.",+ "references": [+ {+ "url": "https://github.com/OpenHands/software-agent-sdk/pull/4796",+ "title": "fix(ci): stop the endpoint audit failing every fork PR",+ "state": "closed"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4769,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4769",+ "title": "[Bug]: Responses streaming overwrites a yielded completion event with stale wrapper state",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:04:33Z",+ "state": "mapped",+ "family": "F06",+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md",+ "feature_ids": [+ "F06.error-events"+ ],+ "basis": "historical-source-reconciliation",+ "reason": "Pinned1.55 captures ResponseCompletedEvent inside sync/async iteration plus completed_response fallback. Original provider stream remains unverified.",+ "reviewed_at": "2026-10-16",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "issue_state": "closed",+ "fresh_live_outcome": null,+ "disposition": "closed-source-changed-awaiting-provider-replay",+ "verification_limit": "October12 evidence is historical, not a runtime verdict on SDK1.55.0/automation1.20.0.",+ "source_refs": [+ "SDK1.55.0: openhands/sdk/llm/llm.py"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4432,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4432",+ "title": "fix(mcp): pin fetch server runtime dependencies",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:03:46Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "Fetch MCP runtime dependency pin request with no concrete Canvas reproduction or user-visible failure in the report.",+ "references": [+ {+ "url": "https://github.com/OpenHands/software-agent-sdk/pull/4303",+ "title": "fix(mcp): pin fetch server runtime dependencies",+ "state": "closed"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4093,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4093",+ "title": "ACP 0.11 drops Gemini model state from NewSessionResponse",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:02:38Z",+ "family": "F13",+ "feature_ids": [+ "F13.acp-conversation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md",+ "issue_state": "closed",+ "previous_state": "mapped",+ "previous_basis": "known-issue",+ "previous_source_target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "outcome": "mapped",+ "basis": "historical-only",+ "current_verdict": "not-run",+ "reason": "Issue closed2026-10-12; inventory has no linked fixing commit. Current main#18131 uses ACP runtime model discovery. Remove wording that presents this as an open report; authentic Gemini NewSessionResponse model-state check needs its real adapter/login and was not performed.",+ "disposition": "closed-without-fresh-authenticated-proof",+ "source_review": {+ "canvas_sources": [+ "src/routes/agent-profiles-settings.tsx",+ "src/routes/agent-settings.tsx",+ "src/components/features/settings/agent-profiles/",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/constants/acp-providers.ts",+ "src/constants/acp-brand-marks.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/hooks/mutation/use-create-conversation.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/settings/acp-conflict-warnings.tsx",+ "src/components/features/settings/acp-credentials-section.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-body.tsx",+ "src/components/features/settings/agent-profiles/agent-profiles-local-view.tsx",+ "src/constants/acp-providers.ts",+ "src/hooks/mutation/use-create-conversation.ts",+ "src/hooks/query/use-acp-model-discovery.ts",+ "src/routes/agent-settings.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "current_map_disposition": "Resolved issue history removed; normal current expectations retained. Acceptance/source evidence remains in this dated report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4265,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4265",+ "title": "OpenHands Fork Audit - MEDIUM: Fork-delta maintenance risk \u2014 carried patches must survive upstream merges",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:12:22Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "Zentra fork patch maintenance, outside this Canvas repository and supported runtime.",+ "references": [],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4264,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4264",+ "title": "OpenHands Fork Audit - HIGH: Stale duplicate build/lib/ directories under vendor/ could be shipped",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:12:07Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "Zentra fork vendor build/lib packaging, outside this Canvas repository and runtime.",+ "references": [],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4263,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4263",+ "title": "OpenHands Fork Audit - HIGH: get_litellm_model_info makes unvalidated httpx.get call at LLM init",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:50:38Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "Zentra fork air-gap/egress policy, outside this Canvas repository and supported runtime.",+ "references": [+ {+ "url": "https://github.com/OpenHands/software-agent-sdk/pull/3912",+ "title": "fix(llm): bound model-info discovery with a timeout to prevent deadlocks",+ "state": "open"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 4248,+ "url": "https://github.com/OpenHands/software-agent-sdk/issues/4248",+ "title": "[Bug]: Missing required parameters for function 'execute_bash': {'security_risk'}",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-12T14:03:08Z",+ "outcome": "out-of-scope",+ "disposition": "closed-out-of-scope",+ "reason": "This unspecified legacy uv CLI report names execute_bash/security_risk but gives no Canvas/Agent Server version, UI entry point or reproduction. The current Canvas launcher is not that SDK CLI; no affected supported Canvas runtime consumer is identified.",+ "references": [+ {+ "url": "https://github.com/OpenHands/OpenHands/pull/14910",+ "title": "test: cover execute_bash without security risk",+ "state": "closed"+ }+ ],+ "verification_limit": "No Canvas runtime fix is claimed for this repository/tooling-only or unsupported consumer report."+ },+ {+ "repository": "OpenHands/automation",+ "number": 551,+ "url": "https://github.com/OpenHands/automation/issues/551",+ "title": "[Bug]: Setting a Git Sync encryption key leaves already-synced automations as plaintext in the repository",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-06T12:10:56Z",+ "issue_state": "closed",+ "family": "F24",+ "feature_ids": [+ "F24.encryption"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F24-git-sync.md",+ "disposition": "accepted-history-current-recheck-pending",+ "pin_status": "accepted-main-or-existing-pin-history",+ "current_result": "not-run",+ "reason": "The frozen Oct10 encryption-key-only re-export check remains dated evidence here. Current F24 keeps key-only set/rotate/clear expectations without the closed 551/fix-PR/date/version proof framing. No new runtime proof on automation 1.20 was performed.",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Existing Oct10 verified encryption-key re-export note is already historical successful behavior, not an active known failure. Keep its date until rechecked on1.20; original closure is automationPR563. Do not infer a new fix verdict from that old run.",+ "map_cleanup": "Resolved issue/history reference removed from the current feature map. Dated evidence remains in this backfill ledger; current expected behavior and distinct open issue prerequisites are retained.",+ "prior_review_reason": "Existing Oct10 verified encryption-key re-export note is already historical successful behavior, not an active known failure. Keep its date until rechecked on1.20; original closure is automationPR563. Do not infer a new fix verdict from that old run."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18181,+ "url": "https://github.com/OpenHands/OpenHands/issues/18181",+ "title": "[Bug]: A failed MCP server save, toggle or delete shows two identical error toasts",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-10T10:21:21Z",+ "issue_state": "closed",+ "family": "F17",+ "feature_ids": [+ "F17.custom-add",+ "F17.enable-disable",+ "F17.delete-server"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "outcome": "mapped",+ "basis": "inherited-main-accepted",+ "disposition": "inherited-main-accepted",+ "current_verdict": "not-run",+ "reason": "Merged main#18182 suppresses duplicate mutation toasts. Map retains one toast per failed action without issue/fix/date history; no additional Oct16 replay claimed.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "current_map_disposition": "Resolved issue history removed; normal current expectations retained. Acceptance/source evidence remains in this dated report."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17958,+ "url": "https://github.com/OpenHands/OpenHands/issues/17958",+ "title": "[Bug]: Custom MCP server editor reports a malformed header as an environment-variable error",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-11T04:07:54Z",+ "issue_state": "closed",+ "family": "F17",+ "feature_ids": [+ "F17.custom-validation"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md",+ "outcome": "mapped",+ "basis": "inherited-main-accepted",+ "disposition": "inherited-main-accepted",+ "current_verdict": "not-run",+ "reason": "Merged main#17966 separates header and environment validation copy. Map retains both current messages without resolved issue history.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": {+ "canvas_sources": [+ "src/routes/extensions-hub.tsx",+ "src/routes/mcp.tsx",+ "src/components/features/skills/extensions-navigation.tsx",+ "src/components/shared/compact-section-navigation.tsx",+ "src/components/features/skills/extensions-mobile-hub.tsx",+ "src/components/features/mcp-page/",+ "src/components/features/settings/mcp-settings/mcp-server-form.tsx",+ "src/utils/mcp-marketplace-utils.ts",+ "src/api/mcp-health/",+ "src/hooks/mutation/use-add-mcp-server.ts",+ "src/utils/mobile-section-nav.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/mcp-page/mcp-server-health.tsx",+ "src/components/features/mcp-page/native-integration-panel.tsx",+ "src/utils/mobile-section-nav.ts"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "current_map_disposition": "Resolved issue history removed; normal current expectations retained. Acceptance/source evidence remains in this dated report."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 17957,+ "url": "https://github.com/OpenHands/OpenHands/issues/17957",+ "title": "[Bug]: Skill detail modal ignores Escape while its '+N' tag popover trigger has focus",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-11T04:02:42Z",+ "issue_state": "closed",+ "family": "F18",+ "feature_ids": [+ "F18.detail-modal"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F18-skills.md",+ "outcome": "mapped",+ "basis": "inherited-main-accepted",+ "disposition": "inherited-main-accepted",+ "current_verdict": "not-run",+ "reason": "Merged main#17967 closes focused tag popover before detail modal on Escape. Map retains that current behavior without resolved issue history.",+ "source_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_review": {+ "canvas_sources": [+ "src/routes/skills-settings.tsx",+ "src/components/features/skills/",+ "src/constants/skills-docs.ts",+ "src/hooks/use-skill-enablement.ts",+ "src/utils/skill-enablement.ts",+ "src/hooks/use-launch-skill-in-chat.ts",+ "src/api/skills-service.ts",+ "src/components/features/chat/skill-install-restart-banner.tsx",+ "src/utils/skill-install-events.ts"+ ],+ "changed_canvas_source_paths": [+ "src/components/features/skills/skill-type-badge.tsx"+ ],+ "issue_status_checked": "2026-10-16",+ "current_pin": "1.55.0",+ "live_claim": "No new runtime claim unless explicitly in checks/current root replay."+ },+ "current_map_refs": [],+ "map_note_present": false,+ "current_map_disposition": "Resolved issue history removed; normal current expectations retained. Acceptance/source evidence remains in this dated report."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18481,+ "url": "https://github.com/OpenHands/OpenHands/issues/18481",+ "title": "[Bug]: Automation setup error pill is below 4.5:1 on Solarized Light (4.33:1) and DeepSea (4.48:1)",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-10-15T15:55:02Z",+ "issue_state": "closed",+ "family": "F28",+ "feature_ids": [+ "F28.error-contrast"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F28-automation-setup-page.md",+ "disposition": "accepted-main-and-current-negative-readback",+ "pin_status": "accepted-main-or-existing-pin-history",+ "current_result": "pass",+ "reason": "The current invalid-cron draft pill uses a 5% red tint. Read and composite actual rendered colors across all five themes: DeepSea 4.672:1, Neutral 4.950:1, Neo 4.950:1, Light+ 4.932:1, Solarized Light 4.623:1. All exceed 4.5:1, supporting retirement of the old 10% pill failure. The Test runs alert was not re-exercised, and no baseline/fix two-by-two proof is claimed.",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "fresh_reproduction_claim": false,+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "state": "not-reproduced",+ "steps": [+ "Create a disposable setup draft named QA_18334_invalid_cron_contrast, prompt \"QA disposable draft. Do not run any tools.\", and Custom cron \"every day\". Save it; for an exact isolated fixture, PATCH /api/automation/v1/drafts/<id> with that draft via control-openhands api --write --data JSON. Validation must say Invalid cron expression and dispatchable=false.",+ "Open /automations/setup?draftId=<id>. Via Settings \u2192 Application \u2192 Color Theme choose each of OpenHands-DeepSea, OpenHands-Neutral, OpenHands-Neo, Light+, Solarized Light and return to the draft. Read computed foreground, pill background alpha and parent surface; composite 5% red(#fb2c36) over the surface and calculate WCAG contrast.",+ "Screenshot each theme. Restore OpenHands-Neutral; do not press Test or dispatch."+ ],+ "artifacts": [+ "evidence/F28.error-contrast/current-deepsea.png",+ "evidence/F28.error-contrast/current-neutral.png",+ "evidence/F28.error-contrast/current-neo.png",+ "evidence/F28.error-contrast/current-light-plus.png",+ "evidence/F28.error-contrast/current-solarized-light.png",+ "evidence/F28.error-contrast/current-colors-with-contrast.json"+ ],+ "basis": "live-check",+ "verification_status": "current-negative",+ "verified_fix": false,+ "post_snapshot_source_caveat": "aa593 changes DeepSea error ink after the tested b34 target. These current-negative numbers remain a b34 readback; current map now carries only the normal 4.5:1 contract. No fresh result for d9ecec is implied.",+ "map_cleanup": "Resolved issue/history reference removed from the current feature map. Dated evidence remains in this backfill ledger; current expected behavior and distinct open issue prerequisites are retained."+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 16559,+ "url": "https://github.com/OpenHands/OpenHands/issues/16559",+ "title": "Automation view takes ~30 seconds to load due to parallel run-history requests",+ "issue_state_at_snapshot": "closed",+ "state_reason": "completed",+ "closed_at": "2026-08-25T11:53:39Z",+ "issue_state": "closed",+ "family": "F21",+ "feature_ids": [+ "F21.card-grid",+ "F21.overview-tiles"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md",+ "disposition": "accepted-history-current-recheck-pending",+ "pin_status": "accepted-main-or-existing-pin-history",+ "current_result": "not-run",+ "reason": "The closed Canvas fan-out incident and Oct12 25-row probe remain historical ledger evidence only. Current F21 keeps the separate still-open automation 347 aggregation/pool/30 s lock issue and its missing original workload prerequisite.",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "fresh_reproduction_claim": false,+ "verification_status": "blocked",+ "prerequisite": "Historical Canvas fan-out incident closed viaPR16867 in August. Keep only historical context beside distinct still-open automation347. Oct12 25-row HTTP200 probe is narrow and does not resolve the backend aggregation/pool/30s-lock workload.",+ "map_cleanup": "Resolved issue/history reference removed from the current feature map. Dated evidence remains in this backfill ledger; current expected behavior and distinct open issue prerequisites are retained.",+ "prior_review_reason": "Historical Canvas fan-out incident closed viaPR16867 in August. Keep only historical context beside distinct still-open automation347. Oct12 25-row HTTP200 probe is narrow and does not resolve the backend aggregation/pool/30s-lock workload."+ }+ ],+ "post_snapshot_updates": [+ {+ "repository": "OpenHands/OpenHands",+ "number": 18324,+ "title": "Use readable theme focus and valid legacy scrim opacity",+ "snapshot_issue_state": "open",+ "issue_state": "closed",+ "closed_at": "2026-10-16T04:12:07Z",+ "state_reason": "completed",+ "family": "F25",+ "feature_ids": [+ "F25.add-backend-modal",+ "F25.agent-server-guidance"+ ],+ "map_file": ".agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md",+ "basis": "source-accepted-after-tested-snapshot",+ "accepted_commit": "508c874ca04edd701298beeaec18bd2924c7827b",+ "accepted_pr": 18327,+ "accepted_main": "d9ecec24181a0dd7b83b5d87561916abb2fe0e8c",+ "included_in_tested_target": false,+ "verification_status": "source-accepted",+ "current_result": "not-run",+ "verified_fix": false,+ "fresh_reproduction_claim": false,+ "reason": "Post-snapshot merge replaces the backend tab, guidance and DigitalOcean provider fixed blue outlines with the theme focus role, preserves outline geometry and host overrides, and strengthens Solarized --oh-focus to #227FBE. The issue closed completed at 04:12:07 UTC. Current map describes the accepted behavior; old failure measurements remain only in backfill data. No live check of this later revision was performed.",+ "secondary_consumer": "F09 exported legacy SettingsMobileDrawer now uses bg-black/50; parent owns its compiled-CSS/current-map reconciliation.",+ "verification_limit": "If runtime fix proof is required, drive the settled keyboard outlines on desktop and phone in all five themes and host --oh-focus override cases on an app build containing 508c874; the existing live checks were on b34ac59."+ }+ ],+ "post_snapshot_source_reviews": [+ {+ "compared_from": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "compared_to": "d9ecec24181a0dd7b83b5d87561916abb2fe0e8c",+ "commits": [+ "aa593c72103862d4f1851988ecacd30f514368dc",+ "508c874ca04edd701298beeaec18bd2924c7827b",+ "d9ecec24181a0dd7b83b5d87561916abb2fe0e8c"+ ],+ "kind": "read-only-source-review",+ "live_rechecked": false,+ "backend_pins_changed": false,+ "tested_live_target_unchanged": true,+ "families": {+ "F25": "Backend focus roles and Solarized focus token updated in accepted source; closed 18324 map failure/history removed, expected contract retained.",+ "F27": "Context meter warning/danger and Compact Context warning now use theme feedback roles; thresholds remain strictly above 70%/90%, API/data flow unchanged. Map now describes the roles and contrast contract.",+ "F28": "Shared DeepSea feedback-error token changes to #EF766A, so prior b34 numeric contrast values are dated ledger evidence only. Current map retains the normal 5% tint and 4.5:1 contract without historical measurements.",+ "F20": "Trust dialog map Gotcha describes current positive focus behavior; old negative/replay history remains in ledger."+ },+ "notes": "d9ecec removes temporary context-color spec bookkeeping only; no runtime behavior added beyond aa593. No UI/merge/product edits by this agent."+ }+ ],+ "checks": [+ {+ "kind": "static-main-css",+ "issue": "OpenHands/OpenHands#18324",+ "consumer": "legacy exported SettingsNavigation/SettingsMobileDrawer (F09 secondary; integrations owns canonical row)",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "source_refs": [+ "src/index.ts",+ "src/lib/index.ts",+ "src/components/settings/index.ts",+ "src/components/features/settings/index.ts",+ "src/components/features/settings/settings-navigation.tsx",+ "src/components/features/settings/settings-mobile-drawer.tsx"+ ],+ "actual": "Real main build/assets/root-CEvL3aWX.css contains .bg-black{background-color:var(--color-black)}, no .bg-opacity-50 utility, and supported slash-opacity utilities. Current routed settings uses SettingsMobileHub.",+ "limit": "Build/source evidence only; no live embedded drawer or visual contrast claim."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5820,+ "feature_ids": [+ "F05.context-window-meter"+ ],+ "result": "blocked",+ "at": "2026-10-16T04:38:33.712264+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "actual": "Actual idle Usage panel has cost0 but usage=null, so no context card/Compact context action. A known gpt-4o dummy model with max_input_tokens128000 and genuine refused-loopback send still yields no context control. Separate direct API diagnostic returned500 No condenser configured; this is not claimed as UI reproduction.",+ "expected": "A reachable Compact context action reports unsupported condenser as a client error.",+ "prerequisite": "Genuine populated token-usage/context metrics on a null/no-op-condenser conversation, requiring successful provider response or an existing recorded real conversation; do not inject fabricated stats.",+ "evidence": [+ "evidence/F05.context-window-meter/condenser-ui-prerequisite-5820.png"+ ],+ "steps": [+ "Create idle tools-empty offline conversation with condenser:null.",+ "Open right panel; choose Usage; inspect absence of Compact context.",+ "Control: known model openai/gpt-4o/max_input_tokens128000, base_url http://127.0.0.1:1/v1, dummy key; send once. Status error and still no context controls.",+ "Separate backend diagnostic only: POST /api/conversations/<idle-id>/condense returned500 No condenser configured."+ ]+ },+ {+ "number": 5790,+ "result": "fail",+ "actual": "Real99-child parent load issued200 batch and opened planner plus parent session sockets. Adding the100th genuine idle child then full reload issued500 batch, displayed Internal Server Error and opened only parent socket, losing discovered planner. No model runs; no browser storage read.",+ "expected": "Planner discovery remains available with100 genuine child IDs; requests obey server limit or fail as an explicit client error.",+ "evidence": [+ "evidence/F06.tool-visualizers/planner-99-control-5790.png",+ "evidence/F06.tool-visualizers/planner-100-batch-5790.png",+ "evidence/F06.tool-visualizers/planner-5790-summary.json"+ ],+ "steps": [+ "Create own tools-empty offline parent and99 idle children with same workspace and real parent_conversation_id; one child tags.plannerparent=<parent>.",+ "In newly initialized configured browser, goto parent; network ids= batch200; lifecycle metadata opens planner and parent.",+ "Create100th native idle child. Full reload resets in-memory planner discovery; no UI-created planner hint was ever arranged.",+ "Actual network batch500; UI toast Internal Server Error; lifecycle metadata shows only new parent connection."+ ],+ "at": "2026-10-16T04:40:32.069176+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F06.tool-visualizers"+ ],+ "known_bug_contract": {+ "id": "F06.tool-visualizers",+ "name": "planner-batch-100",+ "status": "xfail",+ "sha": "e9b66a642e0d7f95",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "500",+ "bugShows": "500",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"500\"",+ "actual": "500",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F06.tool-visualizers-planner-batch-100-1792126521614-1.json"+ }+ },+ {+ "number": 5841,+ "result": "pass",+ "outcome": "out-of-scope",+ "actual": "Actual Canvas Commits lists deleted src/util.py; expanding it displays the defined deleted-file placeholder and sends zero /git/diff requests. Source intentionally disables working-tree D diff. SDK endpoint400 is bypassed by this consumer; no claim server fixed.",+ "expected": "Current F08.diff-deleted-file renders its deleted-file fallback without a failing diff request.",+ "evidence": [+ "evidence/F08.diff-deleted-file/deleted-fallback-5841.png",+ "evidence/F08.diff-deleted-file/deleted-fallback-5841.aria.txt"+ ],+ "steps": [+ "Create disposable Git repo with committed src/util.py, remove only that file, create own idle conversation in repo.",+ "Open right panel \u2192 Commits \u2192 Uncommitted; clear network.",+ "Expand src/util.py via file-diff-viewer-outer containing that path \u2192 collapse button.",+ "Wait exact deleted-file placeholder. browser network --filter /git/diff returns total0."+ ],+ "limit": "No backend-only status proof; this specific SDK route contract has no active working-tree deletion request in Canvas.",+ "at": "2026-10-16T04:42:28.722021+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F08.diff-deleted-file"+ ]+ },+ {+ "number": 5843,+ "result": "fail",+ "actual": "Canvas Commits in repo/src showed mixed paths README.md,new.py,src/app.py,src/util.py. Expanding src/app.py requested repo/src/src/app.py and returned400 twice, so the real file diff could not load. Root repo control showed five correctly based paths.",+ "expected": "Subdirectory workspace lists only files beneath src relative to src and can open their real diff.",+ "evidence": [+ "evidence/F08.diff-view-modes/subdir-mixed-paths-5843.png"+ ],+ "steps": [+ "Owned repo commits README.md,src/app.py,src/util.py; modify README/app, add root notes.txt and src/new.py, remove util.py.",+ "Create idle dummy conversation with working_dir=<repo>/src; open right panel \u2192 Commits \u2192 Uncommitted.",+ "Expand displayed src/app.py; actual /git/diff path is <repo>/src/src/app.py and status400."+ ],+ "at": "2026-10-16T04:43:26.493899+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F08.uncommitted-changes",+ "F08.diff-view-modes"+ ],+ "known_bug_contract": {+ "id": "F08.diff-view-modes",+ "name": "subdirectory-paths",+ "status": "xfail",+ "sha": "2ab83809aa659dcc",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "400",+ "bugShows": "400",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"400\"",+ "actual": "400",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F08.diff-view-modes-subdirectory-paths-1792126467938-1.json"+ }+ },+ {+ "number": 5641,+ "result": "fail",+ "actual": "On pinned1.55.0, actual Commits tab for verified plain parent workspace shows No commits yet; scoped /git/changes returned200 while owned child proj has modified README and untracked notes. Closed upstream fix is not in current pin.",+ "expected": "Changes in the single nested repository remain discoverable.",+ "evidence": [+ "evidence/F08.uncommitted-changes/nested-parent-5641.png"+ ],+ "steps": [+ "Create owned qa-surface-nested/proj Git repo, commit README, then modify it and add notes.txt.",+ "Create idle dummy conversation in plain parent. Verify and remove only its auto-created empty .git (canonical path and zero tracked files).",+ "Reload, open right panel on saved Commits tab. UI No commits yet; /git/changes200; child git status still M README and ??notes."+ ],+ "at": "2026-10-16T04:44:32.149764+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F08.uncommitted-changes"+ ],+ "known_bug_contract": {+ "id": "F08.uncommitted-changes",+ "name": "nested-repository",+ "status": "xfail",+ "sha": "fd68e17f5c9b522e",+ "assertion": {+ "kind": "mismatch",+ "expected": "1",+ "actual": "0",+ "bugShows": "0",+ "bugChecks": [+ {+ "clause": "count eq \"0\"",+ "actual": "0",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F08.uncommitted-changes-nested-repository-1792126580813-1.json"+ }+ },+ {+ "number": 5813,+ "result": "pass",+ "outcome": "out-of-scope",+ "actual": "Actual /btw submission on an owned conversation deleted out of band made a200 batch metadata preflight returning missing conversation, then zero ask_agent requests. fetchConversationData throws V1 conversation not found before the affected route. After metadata polling Canvas returned to list. No assertion that SDK unknown-ID endpoint is fixed.",+ "expected": "Stale /btw must not invoke an unknown conversation as a valid agent.",+ "evidence": [+ "evidence/F05.slash-btw/deleted-conversation-5813.png"+ ],+ "steps": [+ "Open own idle conversation and fill /btw QA_SURFACE stale check.",+ "Delete exactly that owned ID through native API, then press Enter in chat-input.",+ "Inspect scoped network: GET /api/conversations?ids=<deleted>200; ask_agent total0. Source fetchConversationData rejects null preflight before calling askAgent."+ ],+ "limit": "Does not cover deletion in the micro-window between successful preflight and ask_agent; no such controlled concurrent transport fixture was arranged.",+ "at": "2026-10-16T04:46:23.398005+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F05.slash-btw"+ ]+ },+ {+ "number": 5800,+ "result": "fail",+ "actual": "Actual Delete confirmation after out-of-band removal issuedDELETE400 and showed Bad Request. Stale inline rename control PATCH200 was followed by missing metadata and explicit not-found UI, so Canvas does not falsely confirm that rename.",+ "expected": "An already-deleted conversation is reported as missing rather than generic Bad Request; stale rename must not claim success.",+ "evidence": [+ "evidence/F07.delete-confirm/stale-delete-5800.png",+ "evidence/F07.rename-inline/stale-rename-5800.png"+ ],+ "steps": [+ "Open own idle conversation \u2192 header More options \u2192 Delete to leave confirmation open.",+ "Delete same owned ID through native API; click actual Confirm Delete. Network DELETE400; toast Bad Request.",+ "Separate own idle conversation: begin inline rename, delete same ID via API, press Enter. PATCH200, then GET batch returns missing and UI shows not-found without success toast."+ ],+ "limit": "Pause/interrupt variants not driven; only Delete reproduces the generic400 symptom, and rename has a current UI safeguard.",+ "at": "2026-10-16T04:48:05.989823+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F07.delete-confirm",+ "F07.rename-inline"+ ],+ "known_bug_contract": {+ "id": "F07.delete-confirm",+ "name": "already-deleted",+ "status": "xfail",+ "sha": "8c72c361f9674a68",+ "assertion": {+ "kind": "mismatch",+ "expected": "404",+ "actual": "400",+ "bugShows": "400",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"400\"",+ "actual": "400",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F07.delete-confirm-already-deleted-1792126504524-1.json"+ }+ },+ {+ "repository": "OpenHands/OpenHands",+ "number": 18421,+ "result": "fail",+ "actual": "Actual Home New Worktree send posted attachment to original repository path and returned422 path must be inside conversation workspace; created runtime workspace is /tmp/conversation-worktrees/<id>/qa-surface-attachment. Same file sent inside that existing conversation uploaded to worktree path200. Only dummy refused-loopback profile used.",+ "expected": "Home attachment resolves against newly created conversation worktree and uploads200 before message send.",+ "evidence": [+ "evidence/F05.attach-files/home-new-worktree-refresh-18421.png",+ "evidence/F05.attach-files/existing-worktree-control-18421-2.png"+ ],+ "steps": [+ "fixture git-repo --name qa-surface-attachment; fixture file --name qa-surface-note.txt.",+ "Home \u2192 workspace open qa-surface-attachment \u2192 workspace-mode-selector \u2192 new_worktree.",+ "Add file using actual chooser, fill prompt, click Home Send. Network scoped file/upload original-path422; toast preserves detail.",+ "On created conversation add same file and send bounded offline prompt. Upload uses actual worktree path200."+ ],+ "at": "2026-10-16T04:52:27.193836+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "feature_ids": [+ "F05.attach-files"+ ],+ "known_bug_contract": {+ "id": "F05.attach-files",+ "name": "home-new-worktree",+ "status": "xfail",+ "sha": "081b1850a6764519",+ "assertion": {+ "kind": "mismatch",+ "expected": "200",+ "actual": "422",+ "bugShows": "422",+ "bugChecks": [+ {+ "clause": "recent.0.status eq \"422\"",+ "actual": "422",+ "ok": true+ }+ ]+ },+ "doctorBefore": true,+ "doctorAfter": true,+ "transcript": "evidence/known-bugs/F05.attach-files-home-new-worktree-1792126644776-1.json"+ }+ },+ {+ "number": 5630,+ "result": "pass",+ "actual": "Original owned project subagent references missing skill. Actual Home Local Repo launch on1.55 returnedPOST/conversations201, fetched metadata200 and opened conversation route; former missing-skill500 did not occur. Later dummy-model error is separate and expected. Pinned _register_definition catches ValueError, warns and skips invalid agent.",+ "expected": "Bad project subagent does not turn conversation creation into500.",+ "evidence": [+ "evidence/F03.create-error-toast/missing-skill-current-pin-5630.png"+ ],+ "steps": [+ "Create owned .agents/agents/qa-bad-skill.md with name/description and skills: qa-f27-missing-skill.",+ "Keep active explicit refused-loopback QA_SURFACE_REFRESH profile and autotitle off.",+ "Home workspace open owned folder, select Local Repo, fill QA_SURFACE bad-skill pre-inference check, click Home Send.",+ "Actual network create201 and metadata200, URL /conversations/<new-id>. No successful provider response required."+ ],+ "limit": "Verifies creation/pre-inference regression only, not successful model output.",+ "at": "2026-10-16T04:58:19.335138+00:00",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "target": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "sdk": "1.55.0",+ "automation": "1.20.0",+ "repository": "OpenHands/software-agent-sdk",+ "feature_ids": [+ "F03.launch-in-workspace",+ "F03.create-error-toast"+ ]+ },+ {+ "feature_ids": [+ "F10.create-prefill"+ ],+ "repository": "OpenHands/OpenHands",+ "number": 17955,+ "result": "pass",+ "basis": "live-current-acceptance",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "pins": {+ "agentServer": "1.55.0",+ "automation": "1.20.0",+ "agentCanvas": "1.27.1"+ },+ "artifact": "evidence/F10.create-prefill/oct16-inherited-values.json",+ "actual": "UI showed0.4 then provider-linked Save persisted config.temperature0.4 and num_retries0. No validation/model call.",+ "steps": [+ "control-openhands doctor --run <shared18920run>: ok; SDK1.55.0,Canvas1.27.1,automation1.20.0",+ "Set only ownedlauncher-env OPENAI_BASE_URL=OPENAI_API_BASE=http://127.0.0.1:1/v1 and AWS_EC2_METADATA_DISABLED=true; restartwithNode24; doctor; inspect actualserver env booleans true (no key output)",+ "control-openhands api POST /api/profiles/QA_REFRESH_BASE --data '{\"llm\":{\"model\":\"openai/gpt-4o\",\"api_key\":\"<dummy only>\",\"base_url\":\"http://127.0.0.1:1/v1\",\"num_retries\":0,\"temperature\":0.4},\"include_secrets\":true}' --write",+ "control-openhands api POST /api/profiles/QA_REFRESH_BASE/activate --write",+ "control-openhands api POST /api/llm/provider-connections --data '{\"display_name\":\"QA_REFRESH_CONN\",\"provider\":\"openai\",\"api_key\":\"<dummy only>\",\"base_url\":\"http://127.0.0.1:1/v1\"}' --write",+ "await tab.goto('http://127.0.0.1:18920/settings/llm')",+ "await tab.playwright.domSnapshot()",+ "await tab.playwright.getByRole('button',{name:'Add LLM Profile',exact:true}).click()",+ "await tab.playwright.domSnapshot(): All selected, Temperature Optional0.4",+ "await tab.playwright.getByRole('textbox',{name:'Profile Name',exact:true}).fill('QA_REFRESH_INHERIT')",+ "await tab.playwright.getByRole('combobox',{name:'Provider connection',exact:true}).click()",+ "await tab.playwright.getByRole('option',{name:'QA_REFRESH_CONN',exact:true}).click()",+ "await tab.playwright.getByRole('spinbutton',{name:'Temperature Optional',exact:true}).getAttribute('value'): '0.4'",+ "await tab.playwright.getByRole('button',{name:'Save',exact:true}).click()",+ "await tab.playwright.domSnapshot(): Profile QA_REFRESH_INHERIT created; list row present",+ "control-openhands api GET /api/profiles/QA_REFRESH_INHERIT: config.temperature0.4,num_retries0,provider_connection_id matchescreatedconnection"+ ]+ },+ {+ "feature_ids": [+ "F11.row-menu-focus"+ ],+ "repository": "OpenHands/OpenHands",+ "number": 18204,+ "result": "fail",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "actual": "Two real Chrome opens kept activeElement=testid provider-connection-menu-trigger; first ArrowDown kept same focus.",+ "live": {+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "pins": {+ "agentServer": "1.55.0",+ "automation": "1.20.0",+ "agentCanvas": "1.27.1"+ },+ "result": "fail",+ "attempts": 2,+ "observations": [+ {+ "active": "provider-connection-menu-trigger",+ "menu": [+ "Bulk add",+ "Edit",+ "Delete"+ ],+ "afterArrowDown": "provider-connection-menu-trigger"+ },+ {+ "active": "provider-connection-menu-trigger"+ }+ ],+ "steps": [+ "await tab.goto('http://127.0.0.1:18920/settings/llm')",+ "await tab.playwright.getByRole('button',{name:'Provider connection menu',exact:true}).click()",+ "await tab.playwright.evaluate(()=>document.activeElement?.getAttribute('data-testid'))",+ "await tab.playwright.getByRole('menu').innerText()",+ "await tab.playwright.locator(':focus').press('ArrowDown')",+ "await tab.playwright.evaluate(()=>document.activeElement?.getAttribute('data-testid'))",+ "await tab.playwright.locator(':focus').press('Escape')",+ "Repeat menu click/focused-element readback once, then Escape"+ ],+ "arrange": {+ "provider_connection": {+ "display_name": "QA_REFRESH_CONN",+ "provider": "openai",+ "api_key": "<disposable dummy only>",+ "base_url": "http://127.0.0.1:1/v1"+ }+ },+ "limitations": "No provider authentication/inference; this is keyboard-focus behavior."+ }+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5882,+ "feature_ids": [+ "F10.model-picker",+ "F10.add-models"+ ],+ "result": "pass",+ "reason": "API retains Bedrock-only deepseek.* IDs, but current Canvas filters provider options to provider/ names: all3 affected IDs were absent from real Basic and Bulk add. SDK enumeration bug is not a current Canvas picker failure.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root",+ "artifacts": [+ "evidence/F10/5882-current.json",+ "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/evidence/F10.model-picker/oct16-bedrock-catalog.png"+ ],+ "actual": "API retains Bedrock-only deepseek.* IDs, but current Canvas filters provider options to provider/ names: all3 affected IDs were absent from real Basic and Bulk add. SDK enumeration bug is not a current Canvas picker failure."+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5873,+ "feature_ids": [+ "F13.create",+ "F13.edit",+ "F13.set-active"+ ],+ "result": "fail",+ "reason": "Real Canvas create/edit accepted QA_SUFFIX.json; two concurrency-only overwrites each minted a different UUID at revision0. On SDK1.55 the dangling active pointer repaired to default, so no persistent dangling-pointer claim. No model call.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root",+ "artifacts": [+ "evidence/F13/5873-current.json"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5879,+ "feature_ids": [+ "F12.activate"+ ],+ "result": "fail",+ "reason": "With actual global ACP settings, two real Set active clicks each toasted activation success while API/settings active_meta_profile stayed null. Original owned settings restored byte-for-byte; no ACP subprocess or model call.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root",+ "artifacts": [+ "evidence/F12/5879-current.json"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5922,+ "feature_ids": [+ "F12.activate"+ ],+ "result": "fail",+ "reason": "Two real Model Router Set active attempts over owned corrupt settings showed an error toast; activation returned500 with settings-load/path detail instead of409. Corrupt bytes were not overwritten, original bytes restored in finally, doctor healthy. No model call.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root",+ "artifacts": [+ "evidence/F12/5922-current.json"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5897,+ "feature_ids": [+ "F19.refresh"+ ],+ "result": "fail",+ "reason": "Two real plugin Update clicks after moving only the owned source aside surfaced error toasts; refresh returned500 instead of handled400. Installed version/enabled state stayed unchanged. Source restored, own plugin uninstalled and fixture removed; no model call.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root",+ "artifacts": [+ "evidence/F19/5897-current.json"+ ]+ },+ {+ "feature_ids": [+ "F13.llm-pill-precedence",+ "F13.delete"+ ],+ "result": "pass",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root",+ "reason": "Current1.55 runtime confirms per-launch LLM override preserves named profile and active deletion restores default.",+ "steps": [+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/profiles/QA_DRIFT_PIN",+ "--data",+ "@/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/private/root-replays/drift-llm.json",+ "--write"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/agent-profiles/QA_DRIFT",+ "--data",+ "@/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root/private/root-replays/drift-agent.json",+ "--write"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles/QA_DRIFT"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/settings/agents"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-row >> has-text=QA_DRIFT >> testid=agent-profile-menu-trigger"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-set-active"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Switched to profile \"QA_DRIFT\"",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/"+ ],+ [+ "control-openhands",+ "browser",+ "text",+ "testid=chat-input-llm-profile"+ ],+ [+ "control-openhands",+ "conversation",+ "start",+ "--prompt",+ "Reply with exactly qa-profile-check. Do not run tools.",+ "--timeout",+ "30"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/conversations/3e266dbd-ec90-47c2-8938-0e162338e42a"+ ],+ [+ "control-openhands",+ "conversation",+ "pause",+ "3e266dbd-ec90-47c2-8938-0e162338e42a"+ ],+ [+ "control-openhands",+ "browser",+ "goto",+ "/settings/agents"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-row >> has-text=QA_DRIFT >> testid=agent-profile-menu-trigger"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=agent-profile-delete"+ ],+ [+ "control-openhands",+ "browser",+ "click",+ "testid=delete-agent-profile-confirm"+ ],+ [+ "control-openhands",+ "browser",+ "wait-text",+ "Profile \"QA_DRIFT\" deleted",+ "--timeout",+ "10000"+ ],+ [+ "control-openhands",+ "api",+ "GET",+ "/api/agent-profiles"+ ],+ [+ "control-openhands",+ "browser",+ "count",+ "testid=agent-profile-active-badge"+ ],+ [+ "control-openhands",+ "api",+ "POST",+ "/api/agent-profiles/a87f68c9-9109-432a-b0ad-3e255336ab15/activate",+ "--write"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/agent-profiles/QA_DRIFT",+ "--write"+ ],+ [+ "control-openhands",+ "api",+ "DELETE",+ "/api/profiles/QA_DRIFT_PIN",+ "--write"+ ],+ [+ "control-openhands",+ "browser",+ "reload"+ ],+ [+ "control-openhands",+ "doctor"+ ],+ [+ "control-openhands",+ "evidence",+ "add",+ "--feature",+ "F13.llm-pill-precedence",+ "--result",+ "pass",+ "--entry",+ "Oct16 current pin drift",+ "--expected",+ "Named profile fields survive LLM override; active deletion restores valid default",+ "--actual",+ "Actual Home launch used QA_ROOT_BASE openai/gpt-4o while named profile concurrency3/id survived; deleting active QA_DRIFT restored default and one badge. Pre-model only; original profile/pins restored.",+ "--artifact",+ "evidence/F13.llm-pill-precedence/oct16-current-drift.json"+ ],+ [+ "control-openhands",+ "evidence",+ "add",+ "--feature",+ "F13.delete",+ "--result",+ "pass",+ "--entry",+ "Oct16 current pin drift",+ "--expected",+ "Named profile fields survive LLM override; active deletion restores valid default",+ "--actual",+ "Actual Home launch used QA_ROOT_BASE openai/gpt-4o while named profile concurrency3/id survived; deleting active QA_DRIFT restored default and one badge. Pre-model only; original profile/pins restored.",+ "--artifact",+ "evidence/F13.llm-pill-precedence/oct16-current-drift.json"+ ]+ ],+ "observations": {+ "launch": {+ "id": "3e266dbd-ec90-47c2-8938-0e162338e42a",+ "home_llm_pill": {+ "ok": true,+ "text": "QA_ROOT_BASE"+ },+ "profile_pins": "QA_DRIFT_PIN",+ "profile_id": "ddac7abd-ea1a-46a7-80b2-50b15088b8ca",+ "launched_agent_profile": {+ "agent_profile_id": "ddac7abd-ea1a-46a7-80b2-50b15088b8ca",+ "revision": 0,+ "secret_refs": null,+ "llm_profile_ref": "QA_ROOT_BASE"+ },+ "runtime_model": "openai/gpt-4o",+ "runtime_base_url_loopback": true,+ "tool_concurrency_limit": 3+ },+ "delete": {+ "active_profile_name": "default",+ "active_id": "a87f68c9-9109-432a-b0ad-3e255336ab15",+ "badge_count": 1,+ "deleted_profile_absent": true+ }+ },+ "artifacts": [+ "evidence/F13.llm-pill-precedence/oct16-current-drift.json"+ ]+ },+ {+ "repository": "OpenHands/software-agent-sdk",+ "number": 5880,+ "feature_ids": [+ "F10.subscription"+ ],+ "result": "fail",+ "reason": "Two real Connect clicks with the backend routed through an owned refused-loopback proxy returned500 rather than502/504, showing the friendly subscription error plus All connection attempts failed. UV_OFFLINE used the already-cached runtime. Each EXIT restored the original environment and restarted; doctor passed. No subscription account or model call; genuine pending-token poll remains untested.",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-settings-root",+ "artifacts": [+ "evidence/F10.subscription/5880-current.json",+ "evidence/settings-final-restoration.json"+ ],+ "known_bug_attempts": 2+ },+ {+ "kind": "static",+ "family": "F20",+ "command": "control-openhands map check --file F20",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "kind": "static",+ "family": "F21",+ "command": "control-openhands map check --file F21",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "kind": "static",+ "family": "F22",+ "command": "control-openhands map check --file F22",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "kind": "static",+ "family": "F23",+ "command": "control-openhands map check --file F23",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "kind": "static",+ "family": "F24",+ "command": "control-openhands map check --file F24",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "kind": "static",+ "family": "F25",+ "command": "control-openhands map check --file F25",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "kind": "static",+ "family": "F26",+ "command": "control-openhands map check --file F26",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "kind": "static",+ "family": "F27",+ "command": "control-openhands map check --file F27",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "kind": "static",+ "family": "F28",+ "command": "control-openhands map check --file F28",+ "result": "pass",+ "expected": "valid structure, sources, IDs and commands",+ "actual": "ok;28 map files;778 stable IDs;no problems",+ "review_target": "b34ac59e691d16354c14291325d5bd44b37226c5"+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 17564,+ "feature_ids": [+ "F21.phone"+ ],+ "feature": "F21.phone",+ "result": "pass",+ "entry": "Issue OpenHands/OpenHands#17564 current replay",+ "expected": "Heading remains readable without horizontal page overflow.",+ "actual": "Heading 347\u00d728 and 277\u00d728; no page overflow.",+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "Existing user Chrome: Automate \u2192 View mode \u2192 List view. Set 390\u00d7844, then 320\u00d7700 using the viewport capability. Read h1 bounding box, scrollWidth/clientWidth and document width; screenshot both sizes."+ ],+ "artifacts": [+ "evidence/F21.phone/current-390.png",+ "evidence/F21.phone/current-390.json",+ "evidence/F21.phone/current-320.png",+ "evidence/F21.phone/current-320.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 17128,+ "feature_ids": [+ "F21.kebab-menu",+ "F21.phone"+ ],+ "feature": "F21.kebab-menu",+ "result": "pass",+ "entry": "Issue OpenHands/OpenHands#17128 current replay",+ "expected": "Every final-row action stays reachable in the viewport.",+ "actual": "Phone menu x196..356,y510..746; desktop x1248..1408,y666..902. Delete stays inside each.",+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "Automate \u2192 List view. At 390\u00d7844 click the last Automation actions button (scrolls it into view). Read the menu and Delete bounds; close with Escape. Repeat at 1440\u00d71000."+ ],+ "artifacts": [+ "evidence/F21.kebab-menu/current-phone.png",+ "evidence/F21.kebab-menu/current-phone.json",+ "evidence/F21.kebab-menu/current-desktop.png",+ "evidence/F21.kebab-menu/current-desktop.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 16155,+ "feature_ids": [+ "F21.list-row",+ "F21.run-now"+ ],+ "feature": "F21.list-row",+ "result": "pass",+ "entry": "Issue OpenHands/OpenHands#16155 current replay",+ "expected": "Long title leaves Run now compact and readable.",+ "actual": "Run now is icon-only 32\u00d732 px; no document overflow.",+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "At 1440\u00d71000 in Automate \u2192 List view, search \"GitHub PR Reviewer\". Read the exact title and Run now dimensions; leave the disabled automation unexecuted."+ ],+ "artifacts": [+ "evidence/F21.list-row/current-long-title.png",+ "evidence/F21.list-row/current-long-title.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 15833,+ "feature_ids": [+ "F21.add-menu",+ "F22.agent-assisted-setup"+ ],+ "feature": "F21.add-menu",+ "result": "pass",+ "entry": "Issue OpenHands/OpenHands#15833 current replay",+ "expected": "Create opens a usable automation setup interface.",+ "actual": "New Automation form and assistance input visible.",+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "At 1440\u00d71000 click Add Automation \u2192 Create Automation. Wait for the current New Automation setup page and inspect its form and assistance input; do not send a model prompt."+ ],+ "artifacts": [+ "evidence/F22.agent-assisted-setup/current-entry.png",+ "evidence/F22.agent-assisted-setup/current-entry.txt"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/software-agent-sdk",+ "number": 5152,+ "feature_ids": [+ "F27.browser-screenshot"+ ],+ "feature": "F27.browser-screenshot",+ "result": "pass",+ "entry": "Issue OpenHands/software-agent-sdk#5152 current replay",+ "expected": "The installed dependency set imports and constructs the browser executor.",+ "actual": "Exit 0; BrowserToolExecutor; SDK/Agent Server 1.55.0, browser-use 0.11.13, mcp 1.30.0, fastmcp 3.4.8.",+ "steps": [+ "Identify only the owned launcher descendant interpreter: /Users/enyst/.cache/uv/archive-v0/pckcVACfXtx70WBA/bin/python. Set HOME to the run\u2019s private/home and retain closed-loopback OpenAI fallback/AWS metadata protections.",+ "Run that interpreter with: from openhands.tools.browser_use.impl import BrowserToolExecutor; x=BrowserToolExecutor(). Read importlib.metadata.version for openhands-agent-server, openhands-sdk, browser-use, mcp and fastmcp. No browser operation or model request."+ ],+ "artifacts": [+ "evidence/F27.browser-screenshot/current-sdk5152-constructor.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 17563,+ "feature_ids": [+ "F23.phone",+ "F23.edit-close"+ ],+ "feature": "F23.phone",+ "result": "pass",+ "entry": "Issue OpenHands/OpenHands#17563 current replay",+ "expected": "Current edit entry keeps header Save/Close and Name reachable at 390\u00d7844.",+ "actual": "Name/Save/Close all in viewport, dialog count 0, document overflow 0; inner schedule-control overflow is outside this old header-clipping check.",+ "steps": [+ "Arrange 26 disabled real prompt automations with control-openhands api POST /api/automation/v1/preset/prompt --write --data JSON: name QA_18334_integrations_01..25 plus the exact reported long title, prompt \"QA disposable fixture. Do not run any tools. Never dispatch this fixture.\", trigger {type:cron,schedule:\"0 0 1 1 *\",timezone:UTC}, enabled:false. Do not dispatch them.",+ "In Chrome at 390\u00d7844, Automate \u2192 first fixture Automation actions \u2192 Edit. Read Name/Save/Close bounds and dialog count, screenshot, then Close. The old detail Edit and dashboard Edit now share the setup route; this replay used dashboard Edit."+ ],+ "artifacts": [+ "evidence/F23.phone/current-edit-390.png",+ "evidence/F23.phone/current-edit-390.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 18196,+ "feature_ids": [+ "F20.enable-confirm"+ ],+ "feature": "F20.enable-confirm",+ "result": "pass",+ "entry": "Issue OpenHands/OpenHands#18196 current replay",+ "expected": "Trust confirmation contains keyboard focus and restores it to the opener.",+ "actual": "Focus remains in the dialog in both directions; Escape/Cancel restore disabled Enable switch.",+ "steps": [+ "Customize \u2192 Apps \u2192 Add app; install the reviewed repository src/fixtures/canvas-extensions/demo-page disabled.",+ "Focus Enable switch via locator keyboard Space. Read active element; Tab Cancel\u2192Enable, Tab Enable\u2192Cancel, Shift+Tab Cancel\u2192Enable. Escape closes and returns focus to Enable (aria-checked=false).",+ "Space again, Enter on initial Cancel; read restored Enable focus and disabled state. Uninstall only the owned demo through its confirmation. No enable or Update action sent by this drive."+ ],+ "artifacts": [+ "evidence/F20.enable-confirm/current-focus-trap.png",+ "evidence/F20.enable-confirm/current-focus-trap.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 18461,+ "feature_ids": [+ "F28.schedule-presets"+ ],+ "feature": "F28.schedule-presets",+ "result": "fail",+ "entry": "Issue OpenHands/OpenHands#18461 current replay",+ "expected": "Custom preserves Weekly Monday 10:30 as 30 10 * * 1.",+ "actual": "Custom shows 0 9 * * *.",+ "steps": [+ "Automate \u2192 Add Automation \u2192 Create Automation. Choose Weekly (Monday is selected), set At 10:30, then choose Custom. Read automation-setup-custom-schedule. Do not create or dispatch the automation."+ ],+ "artifacts": [+ "evidence/F28.schedule-presets/current-weekly-custom.png",+ "evidence/F28.schedule-presets/current-weekly-custom.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 18481,+ "feature_ids": [+ "F28.error-contrast"+ ],+ "feature": "F28.error-contrast",+ "result": "pass",+ "entry": "Issue OpenHands/OpenHands#18481 current replay",+ "expected": "Invalid-draft pill contrast is at least 4.5:1 in all five themes.",+ "actual": "Minimum 4.623:1 on Solarized Light; every theme exceeds4.5:1.",+ "steps": [+ "Create a disposable setup draft named QA_18334_invalid_cron_contrast, prompt \"QA disposable draft. Do not run any tools.\", and Custom cron \"every day\". Save it; for an exact isolated fixture, PATCH /api/automation/v1/drafts/<id> with that draft via control-openhands api --write --data JSON. Validation must say Invalid cron expression and dispatchable=false.",+ "Open /automations/setup?draftId=<id>. Via Settings \u2192 Application \u2192 Color Theme choose each of OpenHands-DeepSea, OpenHands-Neutral, OpenHands-Neo, Light+, Solarized Light and return to the draft. Read computed foreground, pill background alpha and parent surface; composite 5% red(#fb2c36) over the surface and calculate WCAG contrast.",+ "Screenshot each theme. Restore OpenHands-Neutral; do not press Test or dispatch."+ ],+ "artifacts": [+ "evidence/F28.error-contrast/current-deepsea.png",+ "evidence/F28.error-contrast/current-neutral.png",+ "evidence/F28.error-contrast/current-neo.png",+ "evidence/F28.error-contrast/current-light-plus.png",+ "evidence/F28.error-contrast/current-solarized-light.png",+ "evidence/F28.error-contrast/current-colors-with-contrast.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "run_id": "backfill-18334-refresh",+ "run": "/var/folders/15/pyrd6jhs3tnd259qd39xsplm0000gn/T/openhands-verify/backfill-18334-refresh",+ "source_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "launcher_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "versions": {+ "agent_canvas": "1.27.1",+ "agent_server": "1.55.0",+ "automation": "1.20.0"+ },+ "checked_at": "2026-10-16",+ "fresh_live_check": true,+ "kind": "live",+ "repository": "OpenHands/OpenHands",+ "number": 16217,+ "feature_ids": [+ "F26.runtime-services"+ ],+ "feature": "F26.runtime-services",+ "result": "pass",+ "entry": "Issue OpenHands/OpenHands#16217 current replay",+ "expected": "The current launcher uses the automation environment despite standalone Uvicorn first on PATH.",+ "actual": "Automation health 200 and doctor passes on fixture PATH and restored PATH.",+ "steps": [+ "uv venv --python 3.12 <run>/private/integrations-refresh/uvicorn-only; uv pip install --offline --python <venv>/bin/python uvicorn==0.37.0. Confirm uvicorn version 0.37.0 and importlib.util.find_spec(\"openhands\") is None.",+ "Save <run>/private/launcher-env.json; prepend only <venv>/bin to its PATH. Preserve OPENAI_BASE_URL/OPENAI_API_BASE=http://127.0.0.1:1 and AWS_EC2_METADATA_DISABLED=true. control-openhands restart --timeout 30 --run <run>.",+ "control-openhands doctor --skip-ui and api GET /api/automation/health. Restore original PATH, restart --timeout 30, doctor --skip-ui."+ ],+ "artifacts": [+ "evidence/F26.runtime-services/current-standalone-uvicorn-16217.json",+ "evidence/F26.runtime-services/current-restored-doctor.json"+ ],+ "note": "Current target only; no baseline/fix two-by-two proof claimed. No model requests."+ },+ {+ "name": "OpenHands/OpenHands#17898",+ "feature": "F16.analytics",+ "result": "pass",+ "actual": "Two real opt-in Save/reload cycles on the hard-DNT build produced zero PostHog/telemetry-host requests and no PostHog storage keys. The second reloaded form showed consent checked; consent was then restored off. This covers the hard VITE_DO_NOT_TRACK build, not a browser-only DNT preference.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#4533",+ "feature": "F13.active-drives-conversation",+ "result": "pass",+ "actual": "Four actual seeded-default UI launches retained the dummy LLM key in runtime state and reached only refused-loopback model errors. This verifies propagation, not successful provider authentication.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#4252",+ "feature": "F18.personal-skill-dirs",+ "result": "pass",+ "actual": "A persistence-folder skill appeared in the real Skills page and in initialized conversation events on four UI launches; no successful model reply or historical Docker mount behavior is claimed.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#4709",+ "feature": "F15.agent-context-datetime",+ "result": "pass",+ "actual": "Two UI launches with an explicit2001 settings timestamp produced fresh runtime datetimes and no2001 timestamp in events. Two natural-stale controls also used fresh datetimes. The original setting was restored.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/OpenHands#15694",+ "feature": "F18.toggle",+ "result": "pass",+ "actual": "The owned skill stayed disabled through two eight-second samples and reloads (four aria-checked=false readbacks). This bounded sample does not settle every intermittent timing.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#4891",+ "feature": "F17.agent-uses-server",+ "result": "pass",+ "actual": "Two real initializations with a missing stdio executable and refused HTTP sibling retained the healthy qa_echo tool, persisted user messages and reached the expected loopback model error. Successful provider replies remain outside this check.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#4246",+ "feature": "F17.agent-uses-server",+ "result": "pass",+ "actual": "Two actual90s stdio hang fixtures left initialization after the real MCP timeout, persisted system/user events and reached visible loopback model error. This settles the current Local initialization loop only; successful replies and historical Cloud/SSE remain unverified.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/OpenHands#16578",+ "feature": "F10.validation",+ "result": "pass",+ "actual": "Two genuine Canvas Bedrock starts reached the missing named-profile failure instead of a missing boto3 import. AWS metadata was disabled and both runtime/base endpoints refused loopback. This covers current Local1.55 named-profile import, not AWS credentials, successful inference or Linux fresh installs.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#5882",+ "feature": "F10.model-picker",+ "result": "pass",+ "actual": "API retains Bedrock-only deepseek.* IDs, but current Canvas filters provider options to provider/ names: all3 affected IDs were absent from real Basic and Bulk add. SDK enumeration bug is not a current Canvas picker failure.",+ "run_id": "backfill-18334-settings-root",+ "reason": "API retains Bedrock-only deepseek.* IDs, but current Canvas filters provider options to provider/ names: all3 affected IDs were absent from real Basic and Bulk add. SDK enumeration bug is not a current Canvas picker failure."+ },+ {+ "name": "OpenHands/software-agent-sdk#5873",+ "feature": "F13.create",+ "result": "fail",+ "actual": "Real Canvas create/edit accepted QA_SUFFIX.json; two concurrency-only overwrites each minted a different UUID at revision0. On SDK1.55 the dangling active pointer repaired to default, so no persistent dangling-pointer claim. No model call.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#5879",+ "feature": "F12.activate",+ "result": "fail",+ "actual": "With actual global ACP settings, two real Set active clicks each toasted activation success while API/settings active_meta_profile stayed null. Original owned settings restored byte-for-byte; no ACP subprocess or model call.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#5922",+ "feature": "F12.activate",+ "result": "fail",+ "actual": "Two real Model Router Set active attempts over owned corrupt settings showed an error toast; activation returned500 with settings-load/path detail instead of409. Corrupt bytes were not overwritten, original bytes restored in finally, doctor healthy. No model call.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#5897",+ "feature": "F19.refresh",+ "result": "fail",+ "actual": "Two real plugin Update clicks after moving only the owned source aside surfaced error toasts; refresh returned500 instead of handled400. Installed version/enabled state stayed unchanged. Source restored, own plugin uninstalled and fixture removed; no model call.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/OpenHands#17619",+ "feature": "F06.reconnect",+ "result": "pass",+ "actual": "Two current local trials retained the same open page across an owned backend restart. Each session advanced its after_seq cursor to the prior REST event count minus one, with an observed old-socket close, a new communicating connection and one visible recovery marker. Metadata only: no claim about payload replay, provider streams or Cloud.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#5626",+ "feature": "F06.stop-resume",+ "result": "pass",+ "actual": "During a genuine held dummy-loopback model request, Canvas Stop immediately persisted paused in REST and base_state.json. After SIGKILL of the verified own Agent Server listener and an owned restart, REST/disk and the reloaded Stopped UI retained paused. The bounded accept-only endpoint read no bytes and generated no model reply.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "OpenHands/software-agent-sdk#5880",+ "feature": "F10.subscription",+ "result": "fail",+ "actual": "Two real Connect clicks with the backend routed through an owned refused-loopback proxy returned500 rather than502/504, showing the friendly subscription error plus All connection attempts failed. UV_OFFLINE used the already-cached runtime. Each EXIT restored the original environment and restarted; doctor passed. No subscription account or model call; genuine pending-token poll remains untested.",+ "run_id": "backfill-18334-settings-root"+ },+ {+ "name": "Feature-map structure",+ "result": "pass",+ "actual": "28 families and778 stable IDs; no problems.",+ "command": "control-openhands map check"+ },+ {+ "name": "Route and feature coverage",+ "result": "pass",+ "actual": "31 routes and29 feature component directories accounted for; context-menu is an explicit shared-component exclusion.",+ "command": "control-openhands map coverage"+ },+ {+ "name": "Cited test IDs",+ "result": "pass",+ "actual": "All1103 cited IDs resolve.",+ "command": "control-openhands map testids --strict"+ },+ {+ "name": "Harness unit tests",+ "result": "pass",+ "actual": "64/64 pass under Node24.19.0.",+ "command": "node node_modules/vitest/vitest.mjs run .agents/skills/verify-openhands/scripts/control-openhands.test.mjs"+ },+ {+ "name": "Bounded Python fixture tests",+ "result": "pass",+ "actual": "8/8 pass with the installed SDK Python, including shared-storage and trajectory-overlap fixtures.",+ "command": "python -m unittest discover -s .agents/skills/verify-openhands/scripts/fixtures -p 'test_*.py'"+ },+ {+ "name": "Owned-run teardown",+ "result": "pass",+ "actual": "Both task launchers stopped; every owned port closed; private fixtures purged;117 evidence files retained."+ },+ {+ "name": "Final main application build",+ "result": "pass",+ "actual": "npm run build passed at5385a76fcc9d7ded5cc5fbcdc62b7e464a9136c8, whose product tree matches main d9ecec24181a0dd7b83b5d87561916abb2fe0e8c. Node24.19.0.",+ "command": "npm run build"+ },+ {+ "name": "Final main library build",+ "result": "pass",+ "actual": "npm run build:lib passed at the same merged revision.",+ "command": "npm run build:lib"+ },+ {+ "name": "Final theme and scrim regression tests",+ "result": "pass",+ "actual": "27/27 tests pass in color-themes, settings-navigation and context-meter at the final merged product revision. This checks source/build contracts, not fresh visual fix proof.",+ "command": "vitest run __tests__/themes/color-themes.test.tsx __tests__/components/features/settings/settings-navigation.test.tsx src/components/features/conversation/usage-panel/context-meter.test.tsx"+ },+ {+ "name": "PR scope and ancestry",+ "result": "pass",+ "actual": "All differences from final main are under .agents/skills/verify-openhands/. Both main and the maintainer PR merge are ancestors; original October12 records remain byte-for-byte unchanged."+ }+ ],+ "verification_counts": {+ "blocked": 206,+ "reproduced": 11,+ "resolved-after-snapshot": 1,+ "historical-reproduction": 75,+ "not-reproduced": 18,+ "out-of-scope": 105,+ "not-run": 8+ },+ "publication_inventory_read": {+ "head_sha": "dd08cc1f5e5a989eb1744a8949b1a1ee07153104",+ "head_ref": "docs/feature-map-bug-backfill",+ "head_repo": "OpenHands/OpenHands",+ "base_sha": "aa593c72103862d4f1851988ecacd30f514368dc",+ "main_sha": "d9ecec24181a0dd7b83b5d87561916abb2fe0e8c",+ "main_message": "chore(lint): remove context color spec bookkeeping (#18499)",+ "draft": false,+ "mergeable": true,+ "mergeable_state": "blocked",+ "issue_deltas": {+ "OpenHands": {+ "open": 210,+ "added": [],+ "removed": [+ 18324+ ]+ },+ "software-agent-sdk": {+ "open": 199,+ "added": [],+ "removed": []+ },+ "automation": {+ "open": 14,+ "added": [],+ "removed": []+ }+ }+ },+ "retained_evidence": [+ "evidence/2026-10-16-settings-keyboard.png",+ "evidence/2026-10-16-schedule-custom.png",+ "evidence/2026-10-16-planner-99.png",+ "evidence/2026-10-16-planner-100.png",+ "evidence/2026-10-16-planner-boundary.json",+ "evidence/2026-10-16-reconnect-readback.json",+ "evidence/2026-10-16-pause-durability.json"+ ]+}diff --git a/.agents/skills/verify-openhands/references/backfills/2026-10-16-main-refresh.md b/.agents/skills/verify-openhands/references/backfills/2026-10-16-main-refresh.mdnew file mode 100644index 0000000000..84aa55e02f--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/2026-10-16-main-refresh.md@@ -0,0 +1,584 @@+# Main refresh — 2026-10-16++I am an AI Agent (Codex) helping Engel Nyst (@enyst) with OpenHands project work.++Source reconciled through `main@d9ecec24181a0dd7b83b5d87561916abb2fe0e8c`. Live-check target: `b34ac59e691d16354c14291325d5bd44b37226c5`. Runtime revision: `054521cbff8f676d442f8471983cb2ae7361db2d`, whose product tree matches the live-check target; only this skill differs. Main supplied the accepted maintenance baseline `1c75a05d587bcb502cd3f63d7e13eaa98ca84d53`; this backfill does not advance it.++This reconciles the [frozen October 12 backfill](2026-10-12-open-bugs.md) with main and the later issue inventory. The feature map describes current behavior and removes resolved bug notes and obsolete recipes. This dated ledger retains earlier observations and verification limits. A closed issue, source change or unit test is not a runtime fix proof.++[Current per-issue ledger](2026-10-16-main-refresh.json). [Outcome and label contract](../bug-backfill.md).++## Inventory and outcomes++| Repository | Open bugs | Mapped | Not reproduced | Out of scope | Pending |+| --- | ---: | ---: | ---: | ---: | ---: |+| OpenHands/OpenHands | 211 | 177 | 11 | 23 | 0 |+| OpenHands/software-agent-sdk | 199 | 114 | 7 | 78 | 0 |+| OpenHands/automation | 14 | 10 | 0 | 4 | 0 |++Snapshots are UTC; the report date is October 16 in Europe/Stockholm. All open `bug` issues exclude pull requests. `ready-for-dev` prioritizes checks and does not exclude older unmarked reports.++- OpenHands/OpenHands: `2026-10-15T23:27:24.888074+00:00`.+- OpenHands/software-agent-sdk: `2026-10-15T23:27:28.302757+00:00`.+- OpenHands/automation: `2026-10-15T23:27:30.964305+00:00`.++The refresh adds 96 issues absent from the original snapshot and reconciles 34 original issues now closed, plus 6 further closed reports referenced by main. The 424 open reports at the frozen inventory include 328 retained from October 12. A final inventory read found no additions and one later closure (#18324), leaving 423 open across the three repositories; that later closure is recorded separately below. Canvas has 90 ready-for-dev reports and 121 unmarked reports. The three outcome labels mirror only Canvas-repository accounting; unrelated labels and cross-repository labels are untouched.++## Verification and limits++Three later main commits change theme roles and legacy scrim opacity, without backend pin or API changes. Their source review is recorded separately. Old contrast measurements apply only to the live-check target; the current map contains the accepted contracts.++The real stack uses Canvas 1.27.1, Agent Server/SDK 1.55.0 and automation 1.20.0, Node 24.19.0, native macOS arm64, build `c0efc42e518128f2`. Build and doctor passed. Manual checks use the existing Chrome session; configured harness recipes remain reproducible through `control-openhands`. Two owned runs, `backfill-18334-refresh` (18920) and `backfill-18334-settings-root` (18940), share this build and pins. Manual Chrome is 155.0.8059.82; the configured harness browser is HeadlessChrome 153.0.8010.12. Exact commands and observations are in the JSON ledger.++No successful model, Cloud account, Windows, Docker or Electron behavior is inferred from a loopback fixture. Missing model/platform/account and original-state prerequisites remain explicit. A mapped `Known issue` with an unavailable prerequisite has verification **BLOCKED**; this does not make it a reproduced failure. `Known failure` requires an actual live observation, with its date retained. `not-run` means an available scenario awaits a check. Earlier not-reproduced results apply to their recorded configuration; changed scenarios are rechecked before the current label is retained. This is an issue reconciliation, not a live pass of all 778 feature IDs or a baseline-versus-fix proof.++The combined harness suite passes 64 tests; eight bounded Python fixture tests pass with the installed SDK Python. Map validation, coverage and cited test-ID checks are recorded in the final checks below. Owned runs and fixtures are stopped/cleaned after verification; retained evidence is reviewed before publication.++## Closed-report reconciliation++| Issue | Current disposition | Evidence or remaining prerequisite |+| --- | --- | --- |+| [OpenHands/OpenHands#18481](https://github.com/OpenHands/OpenHands/issues/18481) | accepted-main-and-current-negative-readback | The current invalid-cron draft pill uses a 5% red tint. Read and composite actual rendered colors across all five themes: DeepSea 4.672:1, Neutral 4.950:1, Neo 4.950:1, Light+ 4.932:1, Solarized Light 4.623:1. All exceed 4.5:1, supporting retirement of the old 10% pill failure. The Test runs alert was not re-exercised, and no baseline/fix two-by-two proof is claimed. |+| [OpenHands/OpenHands#18303](https://github.com/OpenHands/OpenHands/issues/18303) | inherited-main-accepted | Merged main#18308/76eadf61f6 uses SETTINGS$SUBSCRIPTION_MODELS_NOT_LOADED in settings and onboarding; upstream map explicitly source-checked2026-10-13. Preserve accepted behavior without a new slow-request UI claim. |+| [OpenHands/OpenHands#18181](https://github.com/OpenHands/OpenHands/issues/18181) | inherited-main-accepted | Merged main#18182 suppresses duplicate mutation toasts. Map retains one toast per failed action without issue/fix/date history; no additional Oct 16 replay claimed. |+| [OpenHands/OpenHands#17958](https://github.com/OpenHands/OpenHands/issues/17958) | inherited-main-accepted | Merged main#17966 separates header and environment validation copy. Map retains both current messages without resolved issue history. |+| [OpenHands/OpenHands#17957](https://github.com/OpenHands/OpenHands/issues/17957) | inherited-main-accepted | Merged main#17967 closes focused tag popover before detail modal on Escape. Map retains that current behavior without resolved issue history. |+| [OpenHands/OpenHands#17955](https://github.com/OpenHands/OpenHands/issues/17955) | live-current-acceptance | Merged main#18422/bf483c12ef persists displayed new-form values. Fresh real Chrome linked create saved inherited temperature 0.4, confirmed GET. Baseline/fix paired known-bug proof was not run, so this is current acceptance, not a new fixed verdict. |+| [OpenHands/OpenHands#17920](https://github.com/OpenHands/OpenHands/issues/17920) | retired-by-current-main | Merged17995/current-main maintenance preserves resolved/rejected groups and normal expectations. Frozen October 12 report retains history; no fresh runtime proof claimed. |+| [OpenHands/OpenHands#17900](https://github.com/OpenHands/OpenHands/issues/17900) | current-keyboard-scenario-pass | On the actual settings page in the existing Chrome session, Escape from Color Theme followed by Tab focused the Analytics checkbox; the next Tab focused Sound and Space toggled it on and off. The settings form returned to its unchanged state. This is a current acceptance observation, not a baseline/fix proof. |+| [OpenHands/OpenHands#17485](https://github.com/OpenHands/OpenHands/issues/17485) | scope-reconfirmed | The reported persistent shared bash-events correlation consumer was removed. Current useBashCommandRunner uses conversation-scoped BashClient.executeCommand. |+| [OpenHands/OpenHands#17096](https://github.com/OpenHands/OpenHands/issues/17096) | closed-out-of-scope | Agent Server version default documentation drift; documentation only. |+| [OpenHands/OpenHands#16748](https://github.com/OpenHands/OpenHands/issues/16748) | closed-out-of-scope | MSW/jsdom teardown unhandled rejection; test runner lifecycle. |+| [OpenHands/OpenHands#16580](https://github.com/OpenHands/OpenHands/issues/16580) | closed-source-changed-awaiting-replay | Touch listener removal now uses capture=true, matching registration. Repeated actual touch drags remain unverified; mouse drag cannot prove this. |+| [OpenHands/OpenHands#16579](https://github.com/OpenHands/OpenHands/issues/16579) | closed-source-changed-awaiting-replay | Active-load markers and loadedConversationId now guard abandoned older-history responses. Delayed-response navigation race needs genuine replay. |+| [OpenHands/OpenHands#16559](https://github.com/OpenHands/OpenHands/issues/16559) | accepted-history-current-recheck-pending | The closed Canvas fan-out incident and Oct 12 25-row probe remain historical ledger evidence only. Current F21 keeps the separate still-open automation 347 aggregation/pool/30 s lock issue and its missing original workload prerequisite. |+| [OpenHands/OpenHands#16429](https://github.com/OpenHands/OpenHands/issues/16429) | closed-out-of-scope | Release-ready workflow permissions; repository CI tooling. |+| [OpenHands/OpenHands#15433](https://github.com/OpenHands/OpenHands/issues/15433) | closed-source-changed-awaiting-replay | Main replaced the streaming presentation path. Successful provider streaming plus an interleaved user message remains unverified. |+| [OpenHands/automation#551](https://github.com/OpenHands/automation/issues/551) | accepted-history-current-recheck-pending | The frozen Oct 10 encryption-key-only re-export check remains dated evidence here. Current F24 keeps key-only set/rotate/clear expectations without the closed 551/fix-PR/date/version proof framing. No new runtime proof on automation 1.20 was performed. |+| [OpenHands/software-agent-sdk#5646](https://github.com/OpenHands/software-agent-sdk/issues/5646) | closed-source-included-runtime-blocked-on-linux | Closed after SDK PR 5759; installed 1.55.0 source has re-probing for live unhealthy backends. No fresh baseline/fix proof. The Oct 12 default stack lacked the app-backend bridge capability; a reviewed platform artifact and admitted bridge deployment are still required. SDK 1.55.0 backend platform selection supports only Linux; this native macOS host cannot execute that fixture. |+| [OpenHands/software-agent-sdk#5641](https://github.com/OpenHands/software-agent-sdk/issues/5641) | closed-upstream-not-in-current-pin | On pinned 1.55.0, actual Commits tab for verified plain parent workspace shows No commits yet; scoped /git/changes returned 200 while owned child proj has modified README and untracked notes. Closed upstream fix is not in current pin. |+| [OpenHands/software-agent-sdk#5640](https://github.com/OpenHands/software-agent-sdk/issues/5640) | closed-upstream-not-in-current-pin | Closed upstream, but pinned 1.55 still uses get_default_config().conversations_path for trajectory ZIP. Deferred-init custom storage remains supported version distinction. |+| [OpenHands/software-agent-sdk#5637](https://github.com/OpenHands/software-agent-sdk/issues/5637) | closed-source-included-runtime-blocked-on-linux | Closed after SDK PR 5774; installed 1.55.0 source has persisted backend process records and startup orphan reaping. No fresh baseline/fix proof. The Oct 12 default stack lacked the app-backend bridge capability; a reviewed platform artifact and admitted bridge deployment are still required. SDK 1.55.0 backend platform selection supports only Linux; this native macOS host cannot execute that fixture. |+| [OpenHands/software-agent-sdk#5630](https://github.com/OpenHands/software-agent-sdk/issues/5630) | closed-fix-in-pin-live-creation-pass | Original owned project subagent references missing skill. Actual Home Local Repo launch on1.55 returnedPOST/conversations201, fetched metadata200 and opened conversation route; former missing-skill500 did not occur. Later dummy-model error is separate and expected. Pinned _register_definition catches ValueError, warns and skips invalid agent. |+| [OpenHands/software-agent-sdk#5581](https://github.com/OpenHands/software-agent-sdk/issues/5581) | closed-awaiting-live-proof | SDK PR5589 closed the report; installed 1.55.0 source uses the cross-process socket creation lock and ownership-aware stale-socket reclamation. Needs a bounded owned shared-TMUX_TMPDIR command fixture for current runtime proof; no current reproduction or fixed verdict is claimed. |+| [OpenHands/software-agent-sdk#5519](https://github.com/OpenHands/software-agent-sdk/issues/5519) | fix-in-current-pin | SDK 1.55 includes c1dfd9c52/#5520, scoped subagents retain conversation-local task_tracker. Current Canvas also still adds task_tracker during scoped profile Save. Remove open Known issue wording, preserving Oct 12 report history; no fresh model-driven delegation proof. |+| [OpenHands/software-agent-sdk#5481](https://github.com/OpenHands/software-agent-sdk/issues/5481) | closed-out-of-scope | Empty bug template identifies no behavior, reproduction, version or mapped Canvas feature; needs a concrete report before Canvas relevance can be established. |+| [OpenHands/software-agent-sdk#5268](https://github.com/OpenHands/software-agent-sdk/issues/5268) | closed-source-changed-awaiting-provider-replay | Pinned normalize_usage guards cache_creation_tokens with model_fields_set and hasattr, defaulting absent/null values to0. Original provider /btw or model response needs replay; source/closure is not live proof. |+| [OpenHands/software-agent-sdk#5213](https://github.com/OpenHands/software-agent-sdk/issues/5213) | closed-source-changed-awaiting-provider-replay | Pinned normalize_usage guards cache_creation_tokens with model_fields_set and hasattr, defaulting absent/null values to0. Original provider /btw or model response needs replay; source/closure is not live proof. |+| [OpenHands/software-agent-sdk#5168](https://github.com/OpenHands/software-agent-sdk/issues/5168) | closed-source-changed-awaiting-provider-replay | Pinned normalize_usage guards cache_creation_tokens with model_fields_set and hasattr, defaulting absent/null values to0. Original provider /btw or model response needs replay; source/closure is not live proof. |+| [OpenHands/software-agent-sdk#5099](https://github.com/OpenHands/software-agent-sdk/issues/5099) | closed-source-changed-awaiting-provider-replay | Pinned normalize_usage guards cache_creation_tokens with model_fields_set and hasattr, defaulting absent/null values to0. Original provider /btw or model response needs replay; source/closure is not live proof. |+| [OpenHands/software-agent-sdk#5072](https://github.com/OpenHands/software-agent-sdk/issues/5072) | closed-out-of-scope | TypeScript Babel override/lockfile consistency for builds; no Canvas runtime feature. |+| [OpenHands/software-agent-sdk#4910](https://github.com/OpenHands/software-agent-sdk/issues/4910) | closed-awaiting-live-proof | SDK 4910 was closed without a merged fixing reference; companion SDK 4901 remains open and maps the same ACP descendant teardown scenario. Retain its exact owned provider-process prerequisite; closure alone is not runtime proof. |+| [OpenHands/software-agent-sdk#4887](https://github.com/OpenHands/software-agent-sdk/issues/4887) | closed-out-of-scope | SDK release publication workflow dispatch; no Canvas feature. |+| [OpenHands/software-agent-sdk#4798](https://github.com/OpenHands/software-agent-sdk/issues/4798) | closed-out-of-scope | TypeScript client endpoint-audit CI comment permissions; no Canvas feature. |+| [OpenHands/software-agent-sdk#4769](https://github.com/OpenHands/software-agent-sdk/issues/4769) | closed-source-changed-awaiting-provider-replay | Pinned1.55 captures ResponseCompletedEvent inside sync/async iteration plus completed_response fallback. Original provider stream remains unverified. |+| [OpenHands/software-agent-sdk#4432](https://github.com/OpenHands/software-agent-sdk/issues/4432) | closed-out-of-scope | Fetch MCP runtime dependency pin request with no concrete Canvas reproduction or user-visible failure in the report. |+| [OpenHands/software-agent-sdk#4265](https://github.com/OpenHands/software-agent-sdk/issues/4265) | closed-out-of-scope | Zentra fork patch maintenance, outside this Canvas repository and supported runtime. |+| [OpenHands/software-agent-sdk#4264](https://github.com/OpenHands/software-agent-sdk/issues/4264) | closed-out-of-scope | Zentra fork vendor build/lib packaging, outside this Canvas repository and runtime. |+| [OpenHands/software-agent-sdk#4263](https://github.com/OpenHands/software-agent-sdk/issues/4263) | closed-out-of-scope | Zentra fork air-gap/egress policy, outside this Canvas repository and supported runtime. |+| [OpenHands/software-agent-sdk#4248](https://github.com/OpenHands/software-agent-sdk/issues/4248) | closed-out-of-scope | This unspecified legacy uv CLI report names execute_bash/security_risk but gives no Canvas/Agent Server version, UI entry point or reproduction. The current Canvas launcher is not that SDK CLI; no affected supported Canvas runtime consumer is identified. |+| [OpenHands/software-agent-sdk#4093](https://github.com/OpenHands/software-agent-sdk/issues/4093) | closed-without-fresh-authenticated-proof | Issue closed2026-10-12; inventory has no linked fixing commit. Current main#18131 uses ACP runtime model discovery. Remove wording that presents this as an open report; authentic Gemini NewSessionResponse model-state check needs its real adapter/login and was not performed. |++## Later closure and retained evidence++[#18324](https://github.com/OpenHands/OpenHands/issues/18324) closed at 2026-10-16T04:12:07Z after the frozen inventory. Main 508c874 adopts theme focus roles and the supported half-opacity legacy scrim; resolved failure notes are removed from F25 and F09. This is source acceptance; no fresh runtime fix proof is claimed for the later theme revision.++[Settings keyboard focus](evidence/2026-10-16-settings-keyboard.png), [schedule switch readback](evidence/2026-10-16-schedule-custom.png), [planner 99 control](evidence/2026-10-16-planner-99.png), [planner 100 failure](evidence/2026-10-16-planner-100.png), and [sanitized planner readback](evidence/2026-10-16-planner-boundary.json). The [reconnect cursor readback](evidence/2026-10-16-reconnect-readback.json) and [pause/crash readback](evidence/2026-10-16-pause-durability.json) retain the final negative-check observations. These reviewed artifacts contain owned fixtures only.++## Frozen open-issue accounting++| Issue | Outcome | Verification | Map | Basis, observation or scope reason |+| --- | --- | --- | --- | --- |+| [OpenHands/OpenHands#18480](https://github.com/OpenHands/OpenHands/issues/18480) | mapped | blocked | [F03.open-repository-dialog](../feature-map/F03-home.md) | Azure DevOps Cloud picker uses direct listing rather than organization installations, omitting other organizations. |+| [OpenHands/OpenHands#18461](https://github.com/OpenHands/OpenHands/issues/18461) | mapped | reproduced | [F28.schedule-presets](../feature-map/F28-automation-setup-page.md) | Fresh create-mode replay still drops Weekly Monday 10:30 when switching to Custom: textbox reads 0 9 * * * instead of 30 10 * * 1. The existing Oct 15 failure remains current on this target. |+| [OpenHands/OpenHands#18445](https://github.com/OpenHands/OpenHands/issues/18445) | mapped | blocked | [F06.load-older-history / F06.reconnect](../feature-map/F06-agent-activity.md) | Cloud initial history failure is swallowed as empty page; successful runtime count can then advance after_seq past unseeded history. |+| [OpenHands/OpenHands#18434](https://github.com/OpenHands/OpenHands/issues/18434) | mapped | blocked | [F25.digitalocean-sign-in](../feature-map/F25-backends-and-cloud.md) | Upstream F25 already records the Oct 14 typed-session-key bridge failure. Current client still reads only baked headers. Requires a separate owned non-baked-key public origin and real API-key-screen entry; the default baked-key stack does not exercise it. |+| [OpenHands/OpenHands#18432](https://github.com/OpenHands/OpenHands/issues/18432) | mapped | blocked | [F14.agent-request](../feature-map/F14-secrets.md) | Current main already records the phone request-secret dialog overflow with an executable known-bug block and a dated2026-10-14 observation. Preserve that date; no Oct 16 reproduction completed. |+| [OpenHands/OpenHands#18421](https://github.com/OpenHands/OpenHands/issues/18421) | mapped | reproduced | [F05.attach-files](../feature-map/F05-composer.md) | Actual Home New Worktree send posted attachment to original repository path and returned 422 path must be inside conversation workspace; created runtime workspace is /tmp/conversation-worktrees/<id>/qa-surface-attachment. Same file sent inside that existing conversation uploaded to worktree path200. Only dummy refused-loopback profile used. |+| [OpenHands/OpenHands#18418](https://github.com/OpenHands/OpenHands/issues/18418) | mapped | blocked | [F03.create-error-toast / F03.launch-with-plugin](../feature-map/F03-home.md) | Hosted1.50.1 runtime fails during skills/plugins setup when bundled plugin-schema is absent. |+| [OpenHands/OpenHands#18390](https://github.com/OpenHands/OpenHands/issues/18390) | mapped | blocked | [F25.add-agent-server / F25.switch-backend](../feature-map/F25-backends-and-cloud.md) | Current-main F25 already preserves historical first-activation race observations; added the current issue link. Recheck at least15 clean/failed-then-clean new-backend activations from a real conversation, observing old-id traffic. A single clean switch cannot clear this intermittent report. |+| [OpenHands/OpenHands#18362](https://github.com/OpenHands/OpenHands/issues/18362) | mapped | blocked | [F26.automation-startup-failure](../feature-map/F26-runtime-variants.md) | Current-main F26 records Oct 13 startup-exit failure and its owned failing-version launcher recipe. Preserve those dates. A new bounded secondary launcher with private HOME/state and free ports is required; it must not disrupt the shared main stack. |+| [OpenHands/OpenHands#18341](https://github.com/OpenHands/OpenHands/issues/18341) | mapped | blocked | [F21.edit / F23.edit-open / F28.kind-toggle](../feature-map/F21-automations-dashboard.md) | Current-main maps the saved-bundle normalization defect in F21 and crosslinks F23/F28. The current form session still drops existingCustomTarballPath. Recheck a reviewed uploaded bundle through dashboard Edit and an unrelated form update, rather than relying on a direct cold URL. |+| [OpenHands/OpenHands#18324](https://github.com/OpenHands/OpenHands/issues/18324) | mapped | resolved-after-snapshot | [F25.add-backend-modal](../feature-map/F25-backends-and-cloud.md) | At the original b34ac59 snapshot, backend tabs/guidance use outline-blue-300. F25 retains Oct 13 focus evidence. F09 secondary consumer is the exported legacy SettingsNavigation/SettingsMobileDrawer build path; the current route uses SettingsMobileHub, so that consumer needs compiled-CSS/export evidence rather than a claimed live phone route. |+| [OpenHands/OpenHands#18302](https://github.com/OpenHands/OpenHands/issues/18302) | mapped | historical-reproduction | [F01.onboarding-repeat-endpoint](../feature-map/F01-first-run-and-sign-in.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#18301](https://github.com/OpenHands/OpenHands/issues/18301) | mapped | historical-reproduction | [F26.session-key-rotated / F26.session-key-persist](../feature-map/F26-runtime-variants.md) | Wrote and read an owned Automation KV marker (HTTP 201/200), then control-openhands restart --rotate-key with no AUTOMATION_KV_SECRET override. Fresh-auth KV GET now returns HTTP 500 instead of the prior marker. |+| [OpenHands/OpenHands#18299](https://github.com/OpenHands/OpenHands/issues/18299) | mapped | historical-reproduction | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Stopped only the owned automation service, backed up its DB, reset alembic_version from 032 to 028 while consecutive_auth_failures exists, restarted the real launcher. Migration duplicate-column signature and automation unavailability occur while launcher stays alive. Private original DB restored afterward. |+| [OpenHands/OpenHands#18287](https://github.com/OpenHands/OpenHands/issues/18287) | mapped | historical-reproduction | [F18.card / F18.facets](../feature-map/F18-skills.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#18234](https://github.com/OpenHands/OpenHands/issues/18234) | mapped | historical-reproduction | [F03.home-run-outcome-health](../feature-map/F03-home.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#18233](https://github.com/OpenHands/OpenHands/issues/18233) | mapped | historical-reproduction | [F25.selector-dropdown](../feature-map/F25-backends-and-cloud.md) | Historical reproduced note dated 2026-10-09 already describes closed-menu Escape clearing the active backend display; source Dropdown and backend selector still own the path. |+| [OpenHands/OpenHands#18210](https://github.com/OpenHands/OpenHands/issues/18210) | mapped | historical-reproduction | [F21.manifest-404](../feature-map/F21-automations-dashboard.md) | Existing missing-manifest recipe and note describe the Home link/toasts; preserved source revision and prior evidence date. |+| [OpenHands/OpenHands#18204](https://github.com/OpenHands/OpenHands/issues/18204) | mapped | reproduced | [F11.row-menu-focus](../feature-map/F11-provider-connections.md) | Reproduced twice in existing Chrome on current pins: opening provider connection menu kept focus on provider-connection-menu-trigger; ArrowDown did not move it. |+| [OpenHands/OpenHands#18196](https://github.com/OpenHands/OpenHands/issues/18196) | not-reproduced | not-reproduced | [F20.enable-confirm](../feature-map/F20-canvas-apps.md) | Fresh keyboard replay on the reviewed local demo app focuses Cancel when trust confirmation opens. Tab and Shift+Tab wrap only between Cancel and Enable trusted app. Escape and keyboard Cancel close it and restore the still-disabled Enable switch. No background action was observed. The old background-focus failure does not reproduce; pending-mutation containment was not exercised. |+| [OpenHands/OpenHands#18161](https://github.com/OpenHands/OpenHands/issues/18161) | mapped | historical-reproduction | [F17.credential-probe](../feature-map/F17-mcp-servers.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#18160](https://github.com/OpenHands/OpenHands/issues/18160) | mapped | historical-reproduction | [F26.frontend-only-returning](../feature-map/F26-runtime-variants.md) | Existing reproduced 2026-10-08 recovery-gate hydrator/toast note preserved. |+| [OpenHands/OpenHands#18155](https://github.com/OpenHands/OpenHands/issues/18155) | mapped | blocked | [F04.show-archived-unarchive](../feature-map/F04-conversation-list.md) | A Cloud sandbox or local runtime that becomes MISSING and non-resumable; ordinary local pause remains resumable. |+| [OpenHands/OpenHands#18136](https://github.com/OpenHands/OpenHands/issues/18136) | mapped | blocked | [F07.cloud-only](../feature-map/F07-conversation-page.md) | An authenticated Cloud sandbox/repository provider and a second fresh browser to compare server repository metadata. |+| [OpenHands/OpenHands#18123](https://github.com/OpenHands/OpenHands/issues/18123) | out-of-scope | out-of-scope | — | Intermittent mock-LLM CI tab-opening wait and diagnostics; report has no reproduction outside the test runner and requests bounded test failure/artifacts. The current user-visible cmd-click feature remains mapped in F04, without claiming the CI flake as a reproduced product bug. |+| [OpenHands/OpenHands#18117](https://github.com/OpenHands/OpenHands/issues/18117) | mapped | blocked | [F06.thinking](../feature-map/F06-agent-activity.md) | A live model that emits a leading inline think block on a tool-call turn, plus its finalized ActionEvent. |+| [OpenHands/OpenHands#18108](https://github.com/OpenHands/OpenHands/issues/18108) | mapped | blocked | [F08.files-tree](../feature-map/F08-workspace-files-and-changes.md) | An authenticated Enterprise Cloud sandbox, no repository selected, with files outside its default working directory. |+| [OpenHands/OpenHands#18063](https://github.com/OpenHands/OpenHands/issues/18063) | mapped | blocked | [F05.overflow-menu](../feature-map/F05-composer.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#18062](https://github.com/OpenHands/OpenHands/issues/18062) | mapped | historical-reproduction | [F21.run-now-error / F23.run-now-error](../feature-map/F23-automation-detail.md) | Existing dashboard and detail stale-automation notes and recipes preserved; no fresh reproduction claimed. |+| [OpenHands/OpenHands#18029](https://github.com/OpenHands/OpenHands/issues/18029) | mapped | blocked | [F25.locked-cloud](../feature-map/F25-backends-and-cloud.md) | Affected locked-to-cloud deployment, authenticated Cloud session and stale-document/alternate-origin evidence are unavailable. |+| [OpenHands/OpenHands#17959](https://github.com/OpenHands/OpenHands/issues/17959) | mapped | blocked | [F22.templates-search / F22.template-card](../feature-map/F22-automation-creation.md) | Both no-result template search and optional-integration pill are already recorded; preserved existing notes. |+| [OpenHands/OpenHands#17953](https://github.com/OpenHands/OpenHands/issues/17953) | mapped | historical-reproduction | [F25.edit-escape / F25.shared-conversation-view / F25.cloud-device-flow](../feature-map/F25-backends-and-cloud.md) | All three reported failures already have existing map notes and drive recipes. |+| [OpenHands/OpenHands#17949](https://github.com/OpenHands/OpenHands/issues/17949) | mapped | historical-reproduction | [F26.host-bind-lan-optin / F26.cli-port-in-use / F26.backend-only](../feature-map/F26-runtime-variants.md) | All three launcher defects already mapped; retained reproduced 2026-10-08 flag note. |+| [OpenHands/OpenHands#17948](https://github.com/OpenHands/OpenHands/issues/17948) | mapped | historical-reproduction | [F23.edit-time-cleared / F23.run-no-conversation / F23.not-found / F23.tarball](../feature-map/F23-automation-detail.md) | All four automation-detail defects already recorded in Gotchas and corresponding recipes. |+| [OpenHands/OpenHands#17946](https://github.com/OpenHands/OpenHands/issues/17946) | mapped | blocked | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Existing npm tmux-isolation known issue preserved; verification launcher intentionally supplies isolation and is not a product-fix claim. |+| [OpenHands/OpenHands#17941](https://github.com/OpenHands/OpenHands/issues/17941) | mapped | not-run | [F18.toggle / F18.toggle-error](../feature-map/F18-skills.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17932](https://github.com/OpenHands/OpenHands/issues/17932) | mapped | not-run | [F15.save-errors](../feature-map/F15-agent-behavior-settings.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17927](https://github.com/OpenHands/OpenHands/issues/17927) | mapped | not-run | [F16.save-states](../feature-map/F16-application-settings.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17922](https://github.com/OpenHands/OpenHands/issues/17922) | mapped | not-run | [F17.install-unsupported](../feature-map/F17-mcp-servers.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17919](https://github.com/OpenHands/OpenHands/issues/17919) | mapped | blocked | [F04.delete-all](../feature-map/F04-conversation-list.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#17899](https://github.com/OpenHands/OpenHands/issues/17899) | mapped | not-run | [F16.git-identity](../feature-map/F16-application-settings.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17898](https://github.com/OpenHands/OpenHands/issues/17898) | not-reproduced | not-reproduced | [F16.analytics](../feature-map/F16-application-settings.md) | Two real opt-in Save/reload cycles on the hard-DNT build produced zero PostHog/telemetry-host requests and no PostHog storage keys. The second reloaded form showed consent checked; consent was then restored off. This covers the hard VITE_DO_NOT_TRACK build, not a browser-only DNT preference. |+| [OpenHands/OpenHands#17872](https://github.com/OpenHands/OpenHands/issues/17872) | out-of-scope | out-of-scope | — | Mock-LLM E2E CI failures and stale test selectors; test harness coverage rather than a proved Canvas user feature failure. |+| [OpenHands/OpenHands#17871](https://github.com/OpenHands/OpenHands/issues/17871) | out-of-scope | out-of-scope | — | Live ACP E2E harness compile-time constant setup; test harness startup rather than the production Canvas path. |+| [OpenHands/OpenHands#17806](https://github.com/OpenHands/OpenHands/issues/17806) | mapped | blocked | [F10.edit](../feature-map/F10-llm-profiles.md) | Current direct-profile Save reached provider preflight, which rejected the refused-loopback dummy endpoint before any profile write. The saved model remained unchanged and the stored key remained present. Key preservation after a completed Save is therefore blocked, not a reproduced failure or a fresh negative. |+| [OpenHands/OpenHands#17803](https://github.com/OpenHands/OpenHands/issues/17803) | mapped | historical-reproduction | [F11.edit / F11.agent-uses-connection](../feature-map/F11-provider-connections.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17801](https://github.com/OpenHands/OpenHands/issues/17801) | mapped | historical-reproduction | [F10.create-prefill / F10.edit](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17788](https://github.com/OpenHands/OpenHands/issues/17788) | mapped | blocked | [F05.attach-files](../feature-map/F05-composer.md) | A Cloud sandbox whose cached runtime URL/key are blank while the fresh runtime lookup is already ready. |+| [OpenHands/OpenHands#17785](https://github.com/OpenHands/OpenHands/issues/17785) | mapped | blocked | [F10.create-validation](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17783](https://github.com/OpenHands/OpenHands/issues/17783) | mapped | historical-reproduction | [F20.git-tree-url / F20.install-error](../feature-map/F20-canvas-apps.md) | Executed both malformed ref and path tree URLs in the live install modal; each raised URI malformed, no install mutation or toast, modal remains. Reviewed local demo control installs and enables. |+| [OpenHands/OpenHands#17774](https://github.com/OpenHands/OpenHands/issues/17774) | mapped | blocked | [F27.planner-empty / F27.planner-plan](../feature-map/F27-workspace-tools.md) | Native Windows local Agent Server and drive-letter workspace fixtures are unavailable. |+| [OpenHands/OpenHands#17773](https://github.com/OpenHands/OpenHands/issues/17773) | mapped | blocked | [F08.files-tree](../feature-map/F08-workspace-files-and-changes.md) | A Windows local Agent Server using cmd.exe; this run is macOS. |+| [OpenHands/OpenHands#17763](https://github.com/OpenHands/OpenHands/issues/17763) | mapped | blocked | [F26.docker-image](../feature-map/F26-runtime-variants.md) | An owned Docker container/daemon is unavailable. |+| [OpenHands/OpenHands#17710](https://github.com/OpenHands/OpenHands/issues/17710) | mapped | blocked | [F14.delete / F14.edit](../feature-map/F14-secrets.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17706](https://github.com/OpenHands/OpenHands/issues/17706) | mapped | blocked | [F10.default-used](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17704](https://github.com/OpenHands/OpenHands/issues/17704) | mapped | historical-reproduction | [F23.script-section](../feature-map/F23-automation-detail.md) | Owned UTF-8 USTAR bundle café.py/main.py, inactive automation python3 café.py: actual Script header renders café.py and no Entrypoint badge; ASCII control untouched. |+| [OpenHands/OpenHands#17657](https://github.com/OpenHands/OpenHands/issues/17657) | out-of-scope | out-of-scope | — | Release-ready workflow permission/startup failure; repository CI tooling. |+| [OpenHands/OpenHands#17623](https://github.com/OpenHands/OpenHands/issues/17623) | mapped | blocked | [F26.helm-chart](../feature-map/F26-runtime-variants.md) | Helm and a disposable Kubernetes cluster are unavailable. |+| [OpenHands/OpenHands#17622](https://github.com/OpenHands/OpenHands/issues/17622) | out-of-scope | out-of-scope | — | Node requirement documentation drift; documentation only. |+| [OpenHands/OpenHands#17621](https://github.com/OpenHands/OpenHands/issues/17621) | mapped | blocked | [F05.context-window-meter](../feature-map/F05-composer.md) | A real summarizing model and conversation whose condensation takes longer than 60 seconds. |+| [OpenHands/OpenHands#17620](https://github.com/OpenHands/OpenHands/issues/17620) | out-of-scope | out-of-scope | — | Fresh-clone typecheck generation order; contributor build tooling. |+| [OpenHands/OpenHands#17619](https://github.com/OpenHands/OpenHands/issues/17619) | not-reproduced | not-reproduced | [F06.reconnect](../feature-map/F06-agent-activity.md) | Two current local trials retained the same open page across an owned backend restart. Each session advanced its after_seq cursor to the prior REST event count minus one, with an observed old-socket close, a new communicating connection and one visible recovery marker. Metadata only: no claim about payload replay, provider streams or Cloud. |+| [OpenHands/OpenHands#17577](https://github.com/OpenHands/OpenHands/issues/17577) | mapped | historical-reproduction | [F10.model-picker](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17567](https://github.com/OpenHands/OpenHands/issues/17567) | mapped | blocked | [F08.diff-view-modes](../feature-map/F08-workspace-files-and-changes.md) | Existing issue link and recorded failure preserved; no fresh reproduction date claimed. |+| [OpenHands/OpenHands#17566](https://github.com/OpenHands/OpenHands/issues/17566) | mapped | historical-reproduction | [F27.terminal-history-reload](../feature-map/F27-workspace-tools.md) | Existing terminal history failure reproduced 2026-10-06 is preserved; no available model credential for a new terminal command. |+| [OpenHands/OpenHands#17564](https://github.com/OpenHands/OpenHands/issues/17564) | not-reproduced | not-reproduced | [F21.phone](../feature-map/F21-automations-dashboard.md) | With 26 disabled automations in the current list view, Dashboard measures 347 px wide at 390×844 and 277 px at 320×700, both 28 px tall, with zero document horizontal overflow. The reported 52 px heading squeeze does not reproduce. |+| [OpenHands/OpenHands#17563](https://github.com/OpenHands/OpenHands/issues/17563) | not-reproduced | not-reproduced | [F23.phone / F23.edit-close](../feature-map/F23-automation-detail.md) | At 390×844, Edit opens the replacement /automations/setup?automationId= page. Save (y65.5..93.5), Close (y65.5..93.5), and Name (y148..184) are inside the viewport; there is no old 940 px modal or negative top offset. This only retires the old modal clipping report: this check does not assess the inner setup schedule-control overflow visible in the screenshot. |+| [OpenHands/OpenHands#17559](https://github.com/OpenHands/OpenHands/issues/17559) | mapped | blocked | [F06.load-older-history](../feature-map/F06-agent-activity.md) | A real persisted history page boundary splitting an action/observation or ACP started/completed pair; dummy-key errors do not produce these tool pairs. |+| [OpenHands/OpenHands#17546](https://github.com/OpenHands/OpenHands/issues/17546) | mapped | historical-reproduction | [F26.cli-port-in-use](../feature-map/F26-runtime-variants.md) | Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. It provides no automatic frontend fallback, and the ingress override does not change that internal port. |+| [OpenHands/OpenHands#17520](https://github.com/OpenHands/OpenHands/issues/17520) | mapped | historical-reproduction | [F10.delete](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17508](https://github.com/OpenHands/OpenHands/issues/17508) | mapped | blocked | [F09.update-available](../feature-map/F09-settings-shell.md) | Isolated old/new package installs with the old process still serving, or the original stale bundle/cache capture after update. |+| [OpenHands/OpenHands#17497](https://github.com/OpenHands/OpenHands/issues/17497) | mapped | blocked | [F04.card](../feature-map/F04-conversation-list.md) | The original corrupted persisted conversation or a deterministic corruption fixture after model-switch failure and fork. |+| [OpenHands/OpenHands#17494](https://github.com/OpenHands/OpenHands/issues/17494) | mapped | blocked | [F07.export-transcript](../feature-map/F07-conversation-page.md) | An authenticated ACP session that emits a titleless tool call, then Markdown and HTML export. |+| [OpenHands/OpenHands#17492](https://github.com/OpenHands/OpenHands/issues/17492) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model emitting failed browser, glob, grep, planning-editor and file-editor observations with is_error=true. |+| [OpenHands/OpenHands#17490](https://github.com/OpenHands/OpenHands/issues/17490) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model producing a completed terminal command with nonzero exit_code and is_error=false. |+| [OpenHands/OpenHands#17476](https://github.com/OpenHands/OpenHands/issues/17476) | mapped | blocked | [F05.context-window-meter](../feature-map/F05-composer.md) | A real planning sub-conversation and main conversation with distinct nonzero usage stats. |+| [OpenHands/OpenHands#17471](https://github.com/OpenHands/OpenHands/issues/17471) | mapped | historical-reproduction | [F05.attach-files](../feature-map/F05-composer.md) | Same-file re-pick after removing its chip leaves attachment count zero. |+| [OpenHands/OpenHands#17470](https://github.com/OpenHands/OpenHands/issues/17470) | mapped | blocked | [F26.docker-image](../feature-map/F26-runtime-variants.md) | Affected Apple Silicon Colima ARM container environment and Docker daemon are unavailable. |+| [OpenHands/OpenHands#17436](https://github.com/OpenHands/OpenHands/issues/17436) | mapped | blocked | [F22.setup-review-confirm](../feature-map/F22-automation-creation.md) | A real LLM key and the reported legacy release/template pair are unavailable. |+| [OpenHands/OpenHands#17429](https://github.com/OpenHands/OpenHands/issues/17429) | mapped | blocked | [F05.slash-btw](../feature-map/F05-composer.md) | The exact pending tool/streaming state that triggers the generator-pickle error, plus a working model for the side question. |+| [OpenHands/OpenHands#17426](https://github.com/OpenHands/OpenHands/issues/17426) | mapped | historical-reproduction | [F13.set-active](../feature-map/F13-agent-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17391](https://github.com/OpenHands/OpenHands/issues/17391) | mapped | blocked | [F10.subscription](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17390](https://github.com/OpenHands/OpenHands/issues/17390) | mapped | blocked | [F25.cloud-sandbox-states](../feature-map/F25-backends-and-cloud.md) | Affected authenticated Cloud account/session and token-provisioning reproduction are unavailable. |+| [OpenHands/OpenHands#17386](https://github.com/OpenHands/OpenHands/issues/17386) | mapped | historical-reproduction | [F04.grouped-folders](../feature-map/F04-conversation-list.md) | Fresh browser puts existing workspace-bound conversations into No workspace. |+| [OpenHands/OpenHands#17384](https://github.com/OpenHands/OpenHands/issues/17384) | mapped | blocked | [F27.launch-child](../feature-map/F27-workspace-tools.md) | A real encrypted provider connection/profile, reviewed file-based worker fixture and authorized model key are unavailable. |+| [OpenHands/OpenHands#17383](https://github.com/OpenHands/OpenHands/issues/17383) | mapped | blocked | [F27.browser-screenshot](../feature-map/F27-workspace-tools.md) | Actual pinned runtime browser-use 0.11.13/mcp 1.30.0/fastmcp 3.4.8 BrowserToolExecutor constructs. The report’s supported Canvas/global browser operation still needs a genuine browser-enabled agent launch and authorized LLM to issue a browser_* operation; a constructor probe alone does not establish that user flow. |+| [OpenHands/OpenHands#17371](https://github.com/OpenHands/OpenHands/issues/17371) | mapped | historical-reproduction | [F25.add-agent-server / F25.recovery-gate](../feature-map/F25-backends-and-cloud.md) | Added QA_Rotation_Legacy for the identical origin through Add Backend using the owned session-key file. After restart --rotate-key and reload, read-only comparison shows default-local matches injected key but the manually registered active entry does not. UI shows Disconnected and Unauthorized; core unreconciled credential persists, although there is visible status feedback in this version. |+| [OpenHands/OpenHands#17271](https://github.com/OpenHands/OpenHands/issues/17271) | mapped | blocked | [F05.context-window-meter](../feature-map/F05-composer.md) | A completed conversation with nonzero token usage before New Chat; a dummy-key run supplies no genuine context usage. |+| [OpenHands/OpenHands#17263](https://github.com/OpenHands/OpenHands/issues/17263) | mapped | blocked | [F05.context-window-meter](../feature-map/F05-composer.md) | A real successful compaction whose condenser usage remains larger than the default agent usage. |+| [OpenHands/OpenHands#17261](https://github.com/OpenHands/OpenHands/issues/17261) | mapped | historical-reproduction | [F10.custom-model / F10.connection-link](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17257](https://github.com/OpenHands/OpenHands/issues/17257) | mapped | blocked | [F04.start-task-cards](../feature-map/F04-conversation-list.md) | An authenticated Cloud backend producing an in-progress sandbox/repository start task. |+| [OpenHands/OpenHands#17239](https://github.com/OpenHands/OpenHands/issues/17239) | mapped | blocked | [F05.llm-profile-picker](../feature-map/F05-composer.md) | October 12 Local eight-profile Home/conversation viewport checks passed. Original multi-row Cloud and real ACP model pickers remain unverified; a Local profile picker is not those consumers. |+| [OpenHands/OpenHands#17233](https://github.com/OpenHands/OpenHands/issues/17233) | mapped | historical-reproduction | [F10.model-picker](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17213](https://github.com/OpenHands/OpenHands/issues/17213) | out-of-scope | out-of-scope | — | The report is OpenHands 0.54.0 runtime-image construction through ToolShield/rootless Podman, including its legacy runtime Docker template; those runtime-builder paths and UI are absent from this Agent Canvas frontend/Agent Server launcher checkout. |+| [OpenHands/OpenHands#17198](https://github.com/OpenHands/OpenHands/issues/17198) | mapped | blocked | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | Affected Windows desktop installation, project definitions and a deterministic failing command fixture are unavailable. |+| [OpenHands/OpenHands#17195](https://github.com/OpenHands/OpenHands/issues/17195) | mapped | historical-reproduction | [F10.connection-link / F10.custom-model](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17163](https://github.com/OpenHands/OpenHands/issues/17163) | mapped | blocked | [F25.cloud-sandbox-states](../feature-map/F25-backends-and-cloud.md) | Affected Free Cloud account, model entitlements and organization history are unavailable. |+| [OpenHands/OpenHands#17162](https://github.com/OpenHands/OpenHands/issues/17162) | mapped | historical-reproduction | [F19.launch-trust](../feature-map/F19-plugins.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17161](https://github.com/OpenHands/OpenHands/issues/17161) | mapped | blocked | [F06.markdown-rendering](../feature-map/F06-agent-activity.md) | A real reasoning, activated-skill, tool-description or error-detail body rendered without includeStandard. The attempted skills dialog exposed public skills only and did not reach the affected chat renderer. |+| [OpenHands/OpenHands#17148](https://github.com/OpenHands/OpenHands/issues/17148) | out-of-scope | out-of-scope | — | Windows CI lane naming/test gating; repository validation workflow. |+| [OpenHands/OpenHands#17135](https://github.com/OpenHands/OpenHands/issues/17135) | mapped | historical-reproduction | [F10.subscription](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17128](https://github.com/OpenHands/OpenHands/issues/17128) | not-reproduced | not-reproduced | [F21.kebab-menu / F21.phone](../feature-map/F21-automations-dashboard.md) | Opened the final (26th) list-row action menu at 390×844 and 1440×1000. Its 160×236 menu and 150×36 Delete button remain fully inside both viewports. No option is clipped in this setup. |+| [OpenHands/OpenHands#17121](https://github.com/OpenHands/OpenHands/issues/17121) | mapped | blocked | [F23.activity-log / F23.run-status-polling](../feature-map/F23-automation-detail.md) | Actual bounded local script run API current_phase=Starting automation, no structured phase fields; Activity Log reads only Running and contains no phase text. |+| [OpenHands/OpenHands#17112](https://github.com/OpenHands/OpenHands/issues/17112) | mapped | blocked | [F26.desktop-main-window / F26.desktop-external-links](../feature-map/F26-runtime-variants.md) | A runnable owned Electron shell and repeatable popup/redirect fixtures are unavailable. |+| [OpenHands/OpenHands#17108](https://github.com/OpenHands/OpenHands/issues/17108) | mapped | blocked | [F03.open-repository-dialog](../feature-map/F03-home.md) | An authenticated Cloud Git provider and a real repository search response that remains pending after clearing the input. |+| [OpenHands/OpenHands#17094](https://github.com/OpenHands/OpenHands/issues/17094) | out-of-scope | out-of-scope | — | Development persistence path documentation drift; documentation only. |+| [OpenHands/OpenHands#17092](https://github.com/OpenHands/OpenHands/issues/17092) | out-of-scope | out-of-scope | — | Session API-key filename documentation drift; documentation only. |+| [OpenHands/OpenHands#17069](https://github.com/OpenHands/OpenHands/issues/17069) | out-of-scope | out-of-scope | — | .env.sample launcher comments; documentation only, with no runtime change requested. |+| [OpenHands/OpenHands#17061](https://github.com/OpenHands/OpenHands/issues/17061) | mapped | historical-reproduction | [F07.rename-inline](../feature-map/F07-conversation-page.md) | Whitespace sends PATCH title empty; current server returns422 and restores the prior title. |+| [OpenHands/OpenHands#17051](https://github.com/OpenHands/OpenHands/issues/17051) | mapped | blocked | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#17026](https://github.com/OpenHands/OpenHands/issues/17026) | mapped | blocked | [F06.thinking](../feature-map/F06-agent-activity.md) | A reasoning-capable live model emitting nonempty MessageEvent.llm_message.reasoning_content on a text-only turn, followed by reload. |+| [OpenHands/OpenHands#16995](https://github.com/OpenHands/OpenHands/issues/16995) | mapped | historical-reproduction | [F20.rail-entry](../feature-map/F20-canvas-apps.md) | Installed and enabled reviewed local demo; hover extension sidebar entry reveals no pin control, while hover Automate reveals sidebar-pin-home-toggle-automations. |+| [OpenHands/OpenHands#16988](https://github.com/OpenHands/OpenHands/issues/16988) | out-of-scope | out-of-scope | — | Mock-LLM E2E reporter success marker after global timeout; test reporting tooling. |+| [OpenHands/OpenHands#16977](https://github.com/OpenHands/OpenHands/issues/16977) | out-of-scope | out-of-scope | — | Mock-LLM E2E reporter counts retry attempts; test reporting tooling. |+| [OpenHands/OpenHands#16954](https://github.com/OpenHands/OpenHands/issues/16954) | mapped | historical-reproduction | [F17.server-health](../feature-map/F17-mcp-servers.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#16951](https://github.com/OpenHands/OpenHands/issues/16951) | mapped | blocked | [F03.composer-blocked-without-llm](../feature-map/F03-home.md) | Current source still gates a named agent on its pinned LLM while launch follows the pill. The older payload fallback symptom is superseded: per-launch LLM override preserves named profile identity/settings, as the fresh F13 check confirms. Map keeps only the unresolved readiness mismatch and the current launch contract. No new credential-presence mismatch trial is claimed. |+| [OpenHands/OpenHands#16950](https://github.com/OpenHands/OpenHands/issues/16950) | mapped | blocked | [F03.selection-after-reload](../feature-map/F03-home.md) | Two independently configured Local backends, or two authenticated Cloud backend/organization scopes, with distinct targets. |+| [OpenHands/OpenHands#16949](https://github.com/OpenHands/OpenHands/issues/16949) | mapped | blocked | [F08.files-refresh](../feature-map/F08-workspace-files-and-changes.md) | An authenticated Cloud repository for the refresh-query mismatch, or real file-editor observations causing an in-place M-to-D Git refetch. |+| [OpenHands/OpenHands#16948](https://github.com/OpenHands/OpenHands/issues/16948) | mapped | historical-reproduction | [F15.save-dirty-state](../feature-map/F15-agent-behavior-settings.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#16947](https://github.com/OpenHands/OpenHands/issues/16947) | mapped | blocked | [F06.load-older-history](../feature-map/F06-agent-activity.md) | A real long conversation whose REST boundary separates one action from its observation; retain the separate companion #17559. |+| [OpenHands/OpenHands#16927](https://github.com/OpenHands/OpenHands/issues/16927) | mapped | historical-reproduction | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Killed only the owned required static frontend after a healthy startup. The genuine launcher remains alive and authenticated /api/settings still returns 200; UI / returns 502 Bad Gateway. This exercises the required-child fatal-exit branch using SIGKILL, not an EADDRINUSE race; exact trigger is recorded. Normal stack restored afterward. |+| [OpenHands/OpenHands#16905](https://github.com/OpenHands/OpenHands/issues/16905) | mapped | blocked | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#16893](https://github.com/OpenHands/OpenHands/issues/16893) | mapped | historical-reproduction | [F18.page](../feature-map/F18-skills.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#16877](https://github.com/OpenHands/OpenHands/issues/16877) | mapped | blocked | [F07.open-by-url](../feature-map/F07-conversation-page.md) | October 12 Local idle-null-title foreground capture showed30-second polling, not reported Cloud terminal-null-title3-second polling. Current Cloud consumer remains mapped. |+| [OpenHands/OpenHands#16876](https://github.com/OpenHands/OpenHands/issues/16876) | mapped | blocked | [F07.history-skeleton](../feature-map/F07-conversation-page.md) | A real planning sub-conversation whose batch lookup fails or yields null, with the main history available. |+| [OpenHands/OpenHands#16874](https://github.com/OpenHands/OpenHands/issues/16874) | mapped | blocked | [F25.cloud-sandbox-states](../feature-map/F25-backends-and-cloud.md) | An authenticated Cloud account, paused sandbox and controlled single-request failure are unavailable. |+| [OpenHands/OpenHands#16872](https://github.com/OpenHands/OpenHands/issues/16872) | mapped | blocked | [F26.desktop-external-links](../feature-map/F26-runtime-variants.md) | A runnable owned Electron shell and inert OAuth authorization URI fixture are unavailable. |+| [OpenHands/OpenHands#16871](https://github.com/OpenHands/OpenHands/issues/16871) | mapped | blocked | [F08.vscode-link](../feature-map/F08-workspace-files-and-changes.md) | An authorized Cloud account/sandbox with a controlled editor exposed_urls value for the current shared-transformer scheme-rejection path. The original card action is absent and the current drawer uses noopener,noreferrer; no exploit or Cloud scheme outcome was reproduced. |+| [OpenHands/OpenHands#16851](https://github.com/OpenHands/OpenHands/issues/16851) | mapped | blocked | [F05.profile-identity](../feature-map/F05-composer.md) | The original conversation/model-switch history and usage record showing Opus with a Fable selector; a dummy-key conversation has no usage evidence. |+| [OpenHands/OpenHands#16844](https://github.com/OpenHands/OpenHands/issues/16844) | mapped | blocked | [F03.workspace-dropdown-search](../feature-map/F03-home.md) | Two independent Local backend workspace registries and parent-directory listings within the cache lifetime. |+| [OpenHands/OpenHands#16842](https://github.com/OpenHands/OpenHands/issues/16842) | mapped | blocked | [F06.reconnect](../feature-map/F06-agent-activity.md) | A transport fixture delaying an old socket close until its replacement is open; ordinary server restart cannot select that ordering. |+| [OpenHands/OpenHands#16839](https://github.com/OpenHands/OpenHands/issues/16839) | mapped | historical-reproduction | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Twenty sequential genuine ingress authenticated /api/conversations/search requests over one keep-alive connection, allHTTP 200; the owned ingress emitted MaxListenersExceededWarning for11 timeout listeners. |+| [OpenHands/OpenHands#16835](https://github.com/OpenHands/OpenHands/issues/16835) | mapped | blocked | [F05.context-window-meter](../feature-map/F05-composer.md) | A working prompt-caching model/proxy and a real conversation reporting nonzero cache_read_tokens. |+| [OpenHands/OpenHands#16802](https://github.com/OpenHands/OpenHands/issues/16802) | mapped | historical-reproduction | [F08.files-rich-plain](../feature-map/F08-workspace-files-and-changes.md) | Rich fenced code renders nested dark/gray containers and padding; layout-only inspection. |+| [OpenHands/OpenHands#16801](https://github.com/OpenHands/OpenHands/issues/16801) | mapped | blocked | [F03.launch-without-workspace](../feature-map/F03-home.md) | A working Claude Opus agent and the original task/prompt that makes it explore the default workspace; the prompt is absent. |+| [OpenHands/OpenHands#16774](https://github.com/OpenHands/OpenHands/issues/16774) | mapped | blocked | [F26.desktop-boot-splash / F26.desktop-main-window](../feature-map/F26-runtime-variants.md) | The reported signed/notarized/quarantined macOS release DMG and an owned installation fixture are unavailable. |+| [OpenHands/OpenHands#16752](https://github.com/OpenHands/OpenHands/issues/16752) | mapped | historical-reproduction | [F04.card-live-status](../feature-map/F04-conversation-list.md) | An empty IDLE conversation uses a green active dot; no pulse animation was observed. |+| [OpenHands/OpenHands#16750](https://github.com/OpenHands/OpenHands/issues/16750) | mapped | blocked | [F04.card](../feature-map/F04-conversation-list.md) | A prompt automation or native delegated task that remains running during conversation-list reload; no usable model is configured here. |+| [OpenHands/OpenHands#16735](https://github.com/OpenHands/OpenHands/issues/16735) | mapped | historical-reproduction | [F19.launch-unicode](../feature-map/F19-plugins.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#16727](https://github.com/OpenHands/OpenHands/issues/16727) | mapped | blocked | [F07.open-by-url](../feature-map/F07-conversation-page.md) | A disposable VM interrupted at the event-write boundary or the original missing-event persisted fixture. |+| [OpenHands/OpenHands#16726](https://github.com/OpenHands/OpenHands/issues/16726) | mapped | historical-reproduction | [F10.model-picker](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#16724](https://github.com/OpenHands/OpenHands/issues/16724) | out-of-scope | out-of-scope | — | Vercel dependency-install Git rewrite configuration; deployment build tooling with no current Canvas runtime reproduction. |+| [OpenHands/OpenHands#16720](https://github.com/OpenHands/OpenHands/issues/16720) | mapped | blocked | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | A bounded reproducible CPU-heavy workload, affected resource envelope and model key are unavailable. |+| [OpenHands/OpenHands#16719](https://github.com/OpenHands/OpenHands/issues/16719) | mapped | blocked | [F06.live-activity](../feature-map/F06-agent-activity.md) | The event/status trace and deterministic stalled or errored state from the original report; its trigger is unspecified. |+| [OpenHands/OpenHands#16716](https://github.com/OpenHands/OpenHands/issues/16716) | mapped | blocked | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | The original simultaneous-agent/shared-repository state or a deterministic stale-diff trigger; the report only suggests that timing. |+| [OpenHands/OpenHands#16690](https://github.com/OpenHands/OpenHands/issues/16690) | mapped | blocked | [F26.docker-image / F26.runtime-services](../feature-map/F26-runtime-variants.md) | Affected Windows-host Docker setup and persisted conversation/workload fixture are unavailable. |+| [OpenHands/OpenHands#16689](https://github.com/OpenHands/OpenHands/issues/16689) | mapped | blocked | [F08.files-load-error](../feature-map/F08-workspace-files-and-changes.md) | The original fresh-VM URL, auth/cookie state and exact request sequence; the report says the trigger is unclear. |+| [OpenHands/OpenHands#16686](https://github.com/OpenHands/OpenHands/issues/16686) | mapped | blocked | [F06.error-events](../feature-map/F06-agent-activity.md) | A Nebius credential and its rejected DeepSeek-V3 request, or the original complete emitted error fixture. |+| [OpenHands/OpenHands#16611](https://github.com/OpenHands/OpenHands/issues/16611) | mapped | blocked | [F22.setup-plugin-create](../feature-map/F22-automation-creation.md) | Authenticated Cloud account, model key and reviewed no-op plugin fixture are unavailable. |+| [OpenHands/OpenHands#16601](https://github.com/OpenHands/OpenHands/issues/16601) | mapped | blocked | [F01.telemetry-consent](../feature-map/F01-first-run-and-sign-in.md) | The multi-day trigger and backend consent response when the modal returns; opted-in localStorage alone does not reproduce it. |+| [OpenHands/OpenHands#16583](https://github.com/OpenHands/OpenHands/issues/16583) | out-of-scope | out-of-scope | — | Issue-readiness and PR-description validation of fenced examples; repository tooling. |+| [OpenHands/OpenHands#16582](https://github.com/OpenHands/OpenHands/issues/16582) | mapped | blocked | [F06.stop-resume](../feature-map/F06-agent-activity.md) | A maximum-iteration legacy error delivered after socket replacement, with a real model run reaching that limit. |+| [OpenHands/OpenHands#16581](https://github.com/OpenHands/OpenHands/issues/16581) | mapped | blocked | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | A supported host fixture with simultaneous independent Terminal consumers and real command histories is unavailable. |+| [OpenHands/OpenHands#16578](https://github.com/OpenHands/OpenHands/issues/16578) | not-reproduced | not-reproduced | [F10.validation](../feature-map/F10-llm-profiles.md) | Two genuine Canvas Bedrock starts reached the missing named-profile failure instead of a missing boto3 import. AWS metadata was disabled and both runtime/base endpoints refused loopback. This covers current Local1.55 named-profile import, not AWS credentials, successful inference or Linux fresh installs. |+| [OpenHands/OpenHands#16551](https://github.com/OpenHands/OpenHands/issues/16551) | mapped | blocked | [F27.browser-screenshot](../feature-map/F27-workspace-tools.md) | Native Windows with both system Chrome and Playwright Chromium plus a model key is unavailable. |+| [OpenHands/OpenHands#16530](https://github.com/OpenHands/OpenHands/issues/16530) | out-of-scope | out-of-scope | — | This is a request to disclose an unspecified security report and request a CVE; it names no Canvas behavior, affected feature, vulnerability or reproducible failure, so there is no feature-map surface to classify. |+| [OpenHands/OpenHands#16526](https://github.com/OpenHands/OpenHands/issues/16526) | mapped | blocked | [F01.onboarding-choose-agent](../feature-map/F01-first-run-and-sign-in.md) | An authenticated hosted Cloud account with the reported organization-membership mismatch during onboarding. |+| [OpenHands/OpenHands#16511](https://github.com/OpenHands/OpenHands/issues/16511) | mapped | blocked | [F26.docker-image](../feature-map/F26-runtime-variants.md) | An owned Docker daemon/image and disposable mount ownership fixture are unavailable. |+| [OpenHands/OpenHands#16506](https://github.com/OpenHands/OpenHands/issues/16506) | out-of-scope | out-of-scope | — | MSW/jsdom ProgressEvent test teardown race; test runner lifecycle. |+| [OpenHands/OpenHands#16492](https://github.com/OpenHands/OpenHands/issues/16492) | mapped | historical-reproduction | [F25.manage-backends / F25.add-agent-server](../feature-map/F25-backends-and-cloud.md) | Read-only same-origin browser inspection records the number of registered backend entries and nonempty plaintext apiKey values; the retained probe now shows both owned entries are readable. No credential value or whole storage JSON was printed. |+| [OpenHands/OpenHands#16455](https://github.com/OpenHands/OpenHands/issues/16455) | mapped | blocked | [F06.events-match](../feature-map/F06-agent-activity.md) | The original duplicated event/stream trace or deterministic prompt and timing; the report provides no reproducible trigger. |+| [OpenHands/OpenHands#16454](https://github.com/OpenHands/OpenHands/issues/16454) | mapped | blocked | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | The original Git changes response and update sequence producing duplicate rows; the report provides no trigger. |+| [OpenHands/OpenHands#16430](https://github.com/OpenHands/OpenHands/issues/16430) | mapped | blocked | [F05.attach-files](../feature-map/F05-composer.md) | A remote or throttled real upload that stays in flight long enough to inspect progress and composer state. |+| [OpenHands/OpenHands#16399](https://github.com/OpenHands/OpenHands/issues/16399) | mapped | historical-reproduction | [F04.pin](../feature-map/F04-conversation-list.md) | A page-two pin disappears after first-page reload although the conversation still exists. |+| [OpenHands/OpenHands#16383](https://github.com/OpenHands/OpenHands/issues/16383) | mapped | historical-reproduction | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Inspect only the owned static-server process for frontend18813: session-key equals a literal argv value. Evidence records booleans only; no credential/argv string exposed. |+| [OpenHands/OpenHands#16374](https://github.com/OpenHands/OpenHands/issues/16374) | out-of-scope | out-of-scope | — | Cloud app-server SQLAlchemy read-session atomicity has no Canvas surface or client-owned contract; implementation is outside this frontend. |+| [OpenHands/OpenHands#16360](https://github.com/OpenHands/OpenHands/issues/16360) | out-of-scope | out-of-scope | — | The issue is labeled Legacy Cloud and describes the removed v0 Cloud composer/status layout, not the current Canvas composer. |+| [OpenHands/OpenHands#16356](https://github.com/OpenHands/OpenHands/issues/16356) | mapped | blocked | [F07.open-by-url](../feature-map/F07-conversation-page.md) | A working provider key that can be rotated/revoked, an existing conversation using it, and a replacement profile. |+| [OpenHands/OpenHands#16341](https://github.com/OpenHands/OpenHands/issues/16341) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model with native code-explorer delegation and the original repository task; no delegated-tool output can be generated without credentials. |+| [OpenHands/OpenHands#16339](https://github.com/OpenHands/OpenHands/issues/16339) | mapped | blocked | [F05.attach-files](../feature-map/F05-composer.md) | An authenticated Cloud upload sandbox and working model to inspect the delivered filename/path notification and read the uploaded file. |+| [OpenHands/OpenHands#16331](https://github.com/OpenHands/OpenHands/issues/16331) | mapped | blocked | [F08.files-tree](../feature-map/F08-workspace-files-and-changes.md) | A real streaming/tool-running conversation while selecting a second workspace file. |+| [OpenHands/OpenHands#16300](https://github.com/OpenHands/OpenHands/issues/16300) | mapped | blocked | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | An uncached isolated reproduction of the reported LiteLLM/Rust dependency toolchain is unavailable. |+| [OpenHands/OpenHands#16298](https://github.com/OpenHands/OpenHands/issues/16298) | mapped | blocked | [F18.skill-in-chat](../feature-map/F18-skills.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#16297](https://github.com/OpenHands/OpenHands/issues/16297) | mapped | historical-reproduction | [F25.selector-dropdown](../feature-map/F25-backends-and-cloud.md) | Served the unchanged UI on an owned protected non-loopback LAN origin, with auth-required and no injected key. Onboarding default Host Name is Local despite the LAN URL; after authenticating with only the private run key and skipping remaining setup, the actual selector reads Connected Local. This verifies the non-loopback naming branch, not a VM installation. |+| [OpenHands/OpenHands#16296](https://github.com/OpenHands/OpenHands/issues/16296) | mapped | blocked | [F13.acp-credentials](../feature-map/F13-agent-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#16277](https://github.com/OpenHands/OpenHands/issues/16277) | mapped | historical-reproduction | [F22.template-card / F22.templates-page](../feature-map/F22-automation-creation.md) | Frozen-source isolated copy with only private @openhands/extensions custom-automation manifest.requires removed; build served correctly from build/ against the healthy owned backend. Actual /automations/templates renders Application Error with TypeError: Cannot read properties of undefined (reading integrations), crashing the entire catalog instead of isolating the malformed entry. Shared source/build untouched. |+| [OpenHands/OpenHands#16263](https://github.com/OpenHands/OpenHands/issues/16263) | mapped | historical-reproduction | [F05.profile-identity](../feature-map/F05-composer.md) | An errored conversation with an unmatched model shows Select a model despite loaded profiles. |+| [OpenHands/OpenHands#16257](https://github.com/OpenHands/OpenHands/issues/16257) | out-of-scope | out-of-scope | — | The issue explicitly scopes an import-time signal-handler ownership/dev-tooling refactor and states there is no user-facing feature or child-service behavior change. No observed Canvas startup/shutdown failure is supplied for the map; concrete required-service failures remain mapped separately in #16927/#18299. |+| [OpenHands/OpenHands#16226](https://github.com/OpenHands/OpenHands/issues/16226) | mapped | historical-reproduction | [F26.cli-port-in-use](../feature-map/F26-runtime-variants.md) | Port 3001 was already occupied; no request or signal was sent to its owner. The genuine frozen launcher with --port 18940 and free private backend/automation ports exits 1 with frontend: port 3001. The explicit ingress override does not choose another frontend port. |+| [OpenHands/OpenHands#16217](https://github.com/OpenHands/OpenHands/issues/16217) | not-reproduced | not-reproduced | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Installed real Uvicorn 0.37.0 alone in an owned Python 3.12.6 venv (no openhands module), prepended its bin only to the private launcher PATH, then restarted the real launcher. Automation 1.20.0 starts, doctor passes, and /api/automation/health returns 200. The reported external-Uvicorn ModuleNotFoundError does not occur in this exact resolver/PATH setup. Normal PATH restored, restart and doctor pass; fallback protections remain. |+| [OpenHands/OpenHands#16210](https://github.com/OpenHands/OpenHands/issues/16210) | mapped | blocked | [F03.plugin-picker](../feature-map/F03-home.md) | An authenticated Cloud organization with configured account plugins before opening New Chat > Plugins. |+| [OpenHands/OpenHands#16208](https://github.com/OpenHands/OpenHands/issues/16208) | mapped | blocked | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | Fedora Rawhide/Python 3.14.6 environment, reported SDK/runtime and model key are unavailable. |+| [OpenHands/OpenHands#16161](https://github.com/OpenHands/OpenHands/issues/16161) | out-of-scope | out-of-scope | — | CI/dependency publication stability audit; contributor build and release tooling. |+| [OpenHands/OpenHands#16155](https://github.com/OpenHands/OpenHands/issues/16155) | not-reproduced | not-reproduced | [F21.list-row / F21.run-now](../feature-map/F21-automations-dashboard.md) | The exact title "GitHub PR Reviewer: OpenHands/software-agent-sdk for @all-hands-bot" appears in a real disabled list row. Its current icon-only Run now button measures 32×32 px with empty text, so the reported Run/now wrapping does not reproduce. |+| [OpenHands/OpenHands#15946](https://github.com/OpenHands/OpenHands/issues/15946) | mapped | blocked | [F04.card](../feature-map/F04-conversation-list.md) | A copied lease-bearing persisted conversation and second owner/expired-lease state; never alter another live run's state. |+| [OpenHands/OpenHands#15833](https://github.com/OpenHands/OpenHands/issues/15833) | not-reproduced | not-reproduced | [F21.add-menu / F22.agent-assisted-setup](../feature-map/F21-automations-dashboard.md) | Add Automation → Create Automation opens /automations/setup with a New Automation editor, Name, Prompt, schedule and a visible "Need help creating an automation?" assistance input. The old modal-to-empty-chat branch no longer reproduces in this entry path. |+| [OpenHands/OpenHands#15822](https://github.com/OpenHands/OpenHands/issues/15822) | mapped | blocked | [F25.switch-backend / F25.health-status / F25.cloud-org-rows](../feature-map/F25-backends-and-cloud.md) | A real registered authenticated Cloud backend/account/organization is unavailable. |+| [OpenHands/OpenHands#15813](https://github.com/OpenHands/OpenHands/issues/15813) | mapped | historical-reproduction | [F15.save-errors](../feature-map/F15-agent-behavior-settings.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#15811](https://github.com/OpenHands/OpenHands/issues/15811) | mapped | historical-reproduction | [F15.verification-page / F15.verification-critic-fields](../feature-map/F15-agent-behavior-settings.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#15782](https://github.com/OpenHands/OpenHands/issues/15782) | mapped | blocked | [F05.draft-persistence](../feature-map/F05-composer.md) | Clarify whether prompt means unsent composer text or the already-sent history message, and capture that state before sidebar Stop. |+| [OpenHands/OpenHands#15778](https://github.com/OpenHands/OpenHands/issues/15778) | mapped | historical-reproduction | [F10.create-validation](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#15701](https://github.com/OpenHands/OpenHands/issues/15701) | mapped | blocked | [F03.launch-in-workspace](../feature-map/F03-home.md) | The original stuck creation request/outcome sequence; the report is intermittent and gives no deterministic trigger. |+| [OpenHands/OpenHands#15694](https://github.com/OpenHands/OpenHands/issues/15694) | not-reproduced | not-reproduced | [F18.toggle](../feature-map/F18-skills.md) | The owned skill stayed disabled through two eight-second samples and reloads (four aria-checked=false readbacks). This bounded sample does not settle every intermittent timing. |+| [OpenHands/OpenHands#15692](https://github.com/OpenHands/OpenHands/issues/15692) | mapped | blocked | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | A real agent commit/push terminal observation while the diff remains mounted, plus a disposable Git remote. |+| [OpenHands/OpenHands#15676](https://github.com/OpenHands/OpenHands/issues/15676) | mapped | blocked | [F25.cloud-sandbox-states](../feature-map/F25-backends-and-cloud.md) | Affected Vercel Cloud session, aged sandbox and exact error/deployment revision are unavailable. |+| [OpenHands/OpenHands#15662](https://github.com/OpenHands/OpenHands/issues/15662) | mapped | historical-reproduction | [F02.sidebar-collapse](../feature-map/F02-app-shell.md) | Collapsed rail has no existing-conversation chooser; expansion is required. |+| [OpenHands/OpenHands#15626](https://github.com/OpenHands/OpenHands/issues/15626) | mapped | blocked | [F06.error-events](../feature-map/F06-agent-activity.md) | The original LiteLLM proxy/database-failure event or a disposable proxy with its PostgreSQL dependency unavailable. |+| [OpenHands/OpenHands#15606](https://github.com/OpenHands/OpenHands/issues/15606) | mapped | blocked | [F06.reload-mid-run](../feature-map/F06-agent-activity.md) | An authenticated Codex ACP session emitting intermediate messages while another window reloads history. |+| [OpenHands/OpenHands#15599](https://github.com/OpenHands/OpenHands/issues/15599) | out-of-scope | out-of-scope | — | The report concerns an agent choosing a Git ref strategy from a free-form task; it identifies no Canvas state, request or Git-control defect. |+| [OpenHands/OpenHands#15567](https://github.com/OpenHands/OpenHands/issues/15567) | mapped | blocked | [F08.files-load-error](../feature-map/F08-workspace-files-and-changes.md) | A plain-HTTP non-loopback deployment and browser accessing that origin, with workspace-cookie observation. |+| [OpenHands/OpenHands#15543](https://github.com/OpenHands/OpenHands/issues/15543) | mapped | blocked | [F07.git-control-bar](../feature-map/F07-conversation-page.md) | The original Git probe stdout/interleaving that temporarily puts a filename in the branch badge; dirty state alone is not a deterministic trigger. |+| [OpenHands/OpenHands#15527](https://github.com/OpenHands/OpenHands/issues/15527) | mapped | blocked | [F06.reconnect](../feature-map/F06-agent-activity.md) | The original custom-domain deployment, idle duration and tab-suspension/network transition; those details are missing. |+| [OpenHands/OpenHands#15519](https://github.com/OpenHands/OpenHands/issues/15519) | mapped | blocked | [F08.files-load-error](../feature-map/F08-workspace-files-and-changes.md) | The reported Linux Docker deployment/auth topology; this isolated run uses the macOS source launcher. |+| [OpenHands/OpenHands#15509](https://github.com/OpenHands/OpenHands/issues/15509) | mapped | blocked | [F22.setup-tarball-create](../feature-map/F22-automation-creation.md) | Native Windows stack and controlled event/webhook delivery fixture are unavailable. |+| [OpenHands/OpenHands#15501](https://github.com/OpenHands/OpenHands/issues/15501) | mapped | blocked | [F08.files-rich-plain](../feature-map/F08-workspace-files-and-changes.md) | The original preview panel and file URL or minimal page identifying the intended surface; Files Rich deliberately blocks scripts, while Browser is separate. |+| [OpenHands/OpenHands#15495](https://github.com/OpenHands/OpenHands/issues/15495) | out-of-scope | out-of-scope | — | The request is for missing AWS Bedrock setup documentation at docs.openhands.dev, with no concrete Canvas control or failing UI behavior. F10 consumes profile forms, not the external documentation authoring surface. |+| [OpenHands/OpenHands#15465](https://github.com/OpenHands/OpenHands/issues/15465) | mapped | blocked | [F23.configuration-event / F23.activity-log](../feature-map/F23-automation-detail.md) | Externally reachable responder, authorized webhook/release repository and controlled real event delivery are unavailable. |+| [OpenHands/OpenHands#15458](https://github.com/OpenHands/OpenHands/issues/15458) | mapped | blocked | [F18.page / F18.project-skill](../feature-map/F18-skills.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#15434](https://github.com/OpenHands/OpenHands/issues/15434) | mapped | blocked | [F26.remote-backend](../feature-map/F26-runtime-variants.md) | Affected non-localhost HTTPS reverse proxy deployment and functioning editor binary are unavailable. |+| [OpenHands/OpenHands#15430](https://github.com/OpenHands/OpenHands/issues/15430) | mapped | blocked | [F17.install-oauth](../feature-map/F17-mcp-servers.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/OpenHands#15421](https://github.com/OpenHands/OpenHands/issues/15421) | mapped | blocked | [F17.agent-uses-server](../feature-map/F17-mcp-servers.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/automation#586](https://github.com/OpenHands/automation/issues/586) | mapped | blocked | [F24.sync-now](../feature-map/F24-git-sync.md) | Automation 1.20 canonical_tarball_bytes still uses tarfile w:gz and compares compressed bytes. Requires a real disposable Git Sync script automation, YAML-only change across different rebuild seconds, and unchanged saved upload identity readback. Serializer-only or decompressed-content equality is not current UI proof. |+| [OpenHands/automation#565](https://github.com/OpenHands/automation/issues/565) | mapped | blocked | [F25.cloud-org-rows / F25.cloud-org-suspended](../feature-map/F25-backends-and-cloud.md) | Authenticated Cloud account/org-bound key and unauthorized-org fixture are unavailable. |+| [OpenHands/automation#557](https://github.com/OpenHands/automation/issues/557) | mapped | historical-reproduction | [F23.configuration / F23.run-linked-conversation](../feature-map/F23-automation-detail.md) | Saved bare qa-backfill-bare agent profile (tools=[], unique persona, closed-loopback qa-offline LLM), created prompt preset then PATCHed its selected agent_profile_id. Real UI Run now created a linked conversation whose launched_agent_profile is null, selected persona is absent and default tools are present; the selected profile was ignored before the deliberate closed-loopback LLM connection failure. |+| [OpenHands/automation#514](https://github.com/OpenHands/automation/issues/514) | mapped | historical-reproduction | [F03.cancel-in-flight-run / F23.activity-log / F23.run-logs-modal](../feature-map/F03-home.md) | Clicked Home owned worker Cancel run; toast Automation run cancelled and Activity Log CANCELLED. The python3 worker.py remains in this run launcher ancestry after cancellation, despite no provider or external worker calls. |+| [OpenHands/automation#472](https://github.com/OpenHands/automation/issues/472) | out-of-scope | out-of-scope | — | Automation subject ownership and cancellation retry contract; Canvas does not submit subject-turn idempotency keys. |+| [OpenHands/automation#470](https://github.com/OpenHands/automation/issues/470) | out-of-scope | out-of-scope | — | Automation subject-turn retry/idempotency API; Canvas does not submit subject turns or idempotency keys in its UI. |+| [OpenHands/automation#418](https://github.com/OpenHands/automation/issues/418) | out-of-scope | out-of-scope | — | PR-review-bot merge gating, review attribution and operational audit; no concrete mapped Canvas UI behavior. |+| [OpenHands/automation#347](https://github.com/OpenHands/automation/issues/347) | mapped | blocked | [F21.card-grid / F21.overview-tiles](../feature-map/F21-automations-dashboard.md) | The owned 25-automation local dashboard completed run summaries with HTTP 200 and no 30 s stall on automation 1.19.3. This narrow live probe does not account for the issue’s missing batch-summary API/pool configuration contracts or withheld original lock workload. A failing dashboard recipe needs the original DB load/lock setup and redacted service-log/HAR timing evidence, linked to Canvas #16559. |+| [OpenHands/automation#343](https://github.com/OpenHands/automation/issues/343) | mapped | historical-reproduction | [F22.setup-tarball-create / F23.script-section / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Uploaded owned tarball scripts/setup.sh writes .qa-setup-marker, main.py requires marker; persisted setup_script_path scripts/setup.sh. UI Run now genuine BashOutput exits1 with QA_NESTED_SETUP_WAS_SKIPPED, proving nested setup ignored. |+| [OpenHands/automation#286](https://github.com/OpenHands/automation/issues/286) | mapped | blocked | [F23.run-status-polling](../feature-map/F23-automation-detail.md) | Owned multi-worker/replica automation deployment and authenticated Cloud sandbox verification context are unavailable. |+| [OpenHands/automation#246](https://github.com/OpenHands/automation/issues/246) | mapped | blocked | [F21.backend-unavailable / F21.list-error](../feature-map/F21-automations-dashboard.md) | The exact 404 URL, user action, version and local launcher/service configuration are missing from the report. |+| [OpenHands/automation#219](https://github.com/OpenHands/automation/issues/219) | mapped | blocked | [F23.run-status-polling](../feature-map/F23-automation-detail.md) | Owned disposable Cloud account at its sandbox cap with an interactive fixture is unavailable. |+| [OpenHands/automation#200](https://github.com/OpenHands/automation/issues/200) | out-of-scope | out-of-scope | — | Automation KV REST request strictness and response contracts; no Canvas KV editor or consumer of these writes. |+| [OpenHands/automation#113](https://github.com/OpenHands/automation/issues/113) | mapped | blocked | [F23.run-linked-conversation](../feature-map/F23-automation-detail.md) | Affected automation/run/conversation ids, deployment version and genuine plugin-run credentials are unavailable. |+| [OpenHands/software-agent-sdk#5995](https://github.com/OpenHands/software-agent-sdk/issues/5995) | mapped | blocked | [F13.acp-conversation / F13.acp-model-discovery](../feature-map/F13-agent-profiles.md) | Current SDK 1.55 retains Claude ACP0.63.0, the version named by the report. Canvas uses SDK registry presets and live discovery, so first-turn Opus5.5 rejection is an actual consumer. No authenticated Claude turn was attempted. |+| [OpenHands/software-agent-sdk#5985](https://github.com/OpenHands/software-agent-sdk/issues/5985) | mapped | blocked | [F19.refresh / F19.enable-toggle](../feature-map/F19-plugins.md) | Canvas Update calls the shared plugin refresh endpoint and listing can run concurrently. Disabled-state/source loss during directory replacement is therefore relevant; the issue forces copytree scheduling in Python, which is not live Canvas proof. |+| [OpenHands/software-agent-sdk#5974](https://github.com/OpenHands/software-agent-sdk/issues/5974) | mapped | blocked | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | An owned workspace-registry lock plus a waiting workspace mutation can block server liveness and Canvas. Current async workspaces routes call blocking store.update. Requires a bounded lock-holder with ownership/wait confirmation, concurrent liveness probes, release and successful readback; the original exact waiter observation uses Linux /proc/locks. |+| [OpenHands/software-agent-sdk#5969](https://github.com/OpenHands/software-agent-sdk/issues/5969) | mapped | blocked | [F06.stop-resume](../feature-map/F06-agent-activity.md) | Canvas Run/Play and send run:true consume the lifecycle that can report finished yet reject Run or delay follow-up while webhook is pending. |+| [OpenHands/software-agent-sdk#5963](https://github.com/OpenHands/software-agent-sdk/issues/5963) | out-of-scope | out-of-scope | — | Automation presets instantiate Python RemoteConversation but do not enumerate its state.events or compare that cache count with the server: they use their received_events callback list. Canvas uses TypeScript event stores. This report specifies extra transient deltas in Python state.events; no concrete affected Canvas widget/run outcome for that exact cache contract was identified. |+| [OpenHands/software-agent-sdk#5962](https://github.com/OpenHands/software-agent-sdk/issues/5962) | out-of-scope | out-of-scope | — | OpenAI-compatible /v1/chat/completions continued-conversation gateway status handling. Canvas creates/sends conversations through ConversationClient, not the OpenAI gateway. |+| [OpenHands/software-agent-sdk#5961](https://github.com/OpenHands/software-agent-sdk/issues/5961) | out-of-scope | out-of-scope | — | Agent Server PostHog outcome token/cost bucket instrumentation. Canvas renders conversation metrics and has its own telemetry owner; it does not consume these backend analytics payloads. |+| [OpenHands/software-agent-sdk#5960](https://github.com/OpenHands/software-agent-sdk/issues/5960) | out-of-scope | out-of-scope | — | Agent Server TelemetrySpec.salt OpenAPI description/fallback documentation; Canvas exposes consent but no TelemetrySpec salt field. No user-facing Canvas contract change in this report. |+| [OpenHands/software-agent-sdk#5959](https://github.com/OpenHands/software-agent-sdk/issues/5959) | out-of-scope | out-of-scope | — | Python Conversation(token_callbacks=...) forwarding into RemoteConversation. Canvas receives tokens over the TypeScript WebSocket/event path and does not instantiate this Python callback API. |+| [OpenHands/software-agent-sdk#5952](https://github.com/OpenHands/software-agent-sdk/issues/5952) | out-of-scope | out-of-scope | — | Agent Server outbound event WebhookSpec batch/flush scheduling, not Canvas automation inbound custom-webhook registration. No mapped Canvas setting/consumer for this receiver batching contract. |+| [OpenHands/software-agent-sdk#5947](https://github.com/OpenHands/software-agent-sdk/issues/5947) | out-of-scope | out-of-scope | — | Canvas own status mapper accepts uppercase wire and lowercase statuses for Local and Cloud, not erroneous exported GitChange type comparisons. |+| [OpenHands/software-agent-sdk#5946](https://github.com/OpenHands/software-agent-sdk/issues/5946) | out-of-scope | out-of-scope | — | Canvas ignores PATCH result then batch-fetches canonical conversation. It does not read id/title from mis-typed update response. |+| [OpenHands/software-agent-sdk#5941](https://github.com/OpenHands/software-agent-sdk/issues/5941) | out-of-scope | out-of-scope | — | OpenAI-compatible /v1/chat/completions concurrent continued requests. Canvas uses ConversationClient send/run endpoints, not this gateway. |+| [OpenHands/software-agent-sdk#5940](https://github.com/OpenHands/software-agent-sdk/issues/5940) | out-of-scope | out-of-scope | — | Requires Python client-local custom tool registration and tool_module_qualnames import mismatch. Canvas encrypted-settings/profile creation does not submit that module registration contract. |+| [OpenHands/software-agent-sdk#5939](https://github.com/OpenHands/software-agent-sdk/issues/5939) | out-of-scope | out-of-scope | — | Agent Server analytics deferred_init property on warm-pool pods. Canvas does not render or consume this backend telemetry field. |+| [OpenHands/software-agent-sdk#5938](https://github.com/OpenHands/software-agent-sdk/issues/5938) | mapped | blocked | [F23.run-status-polling / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Automation1.20 prompt/plugin sdk_main callbacks append received events and publish ActionEvent.summary as live phase. A real socket reconnect can omit actions recorded during the gap; completion metadata loss is not established. Prerequisite: an owned remote continuation, genuine tool events across a controlled reconnect and the phase consumer visible; separate current phase defect17121 may mask this symptom. |+| [OpenHands/software-agent-sdk#5937](https://github.com/OpenHands/software-agent-sdk/issues/5937) | mapped | blocked | [F23.run-status-polling / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Automation1.20 prompt/plugin presets attach AUTOMATION_CONVERSATION_ID, send the user prompt and call synchronous RemoteConversation.run(). Continuing an owned persisted conversation already running a genuine tool longer than30s can hit the SDK trigger read timeout while the run is healthy. Prerequisite: controlled continued automation plus its real long-running model/tool turn; a stopped dummy provider is not this join scenario. |+| [OpenHands/software-agent-sdk#5936](https://github.com/OpenHands/software-agent-sdk/issues/5936) | out-of-scope | out-of-scope | — | WebhookSpec description of outbound webhook event path/flush timer. Documentation-only backend contract, distinct from Canvas automation custom inbound webhooks. |+| [OpenHands/software-agent-sdk#5935](https://github.com/OpenHands/software-agent-sdk/issues/5935) | mapped | blocked | [F20.app-backend-view](../feature-map/F20-canvas-apps.md) | Canvas consumes app_backend_ingress_url/canvas_app_backend_bridge_v1; an environment null does not clear a JSON-config public URL. Recheck an admitted bridge-capable server with those conflicting sources and the actual app-backend page; default launcher configuration does not arrange this state. |+| [OpenHands/software-agent-sdk#5934](https://github.com/OpenHands/software-agent-sdk/issues/5934) | out-of-scope | out-of-scope | — | OpenAI-compatible /v1/chat/completions continued-conversation retry persistence. Canvas sends user messages through ConversationClient rather than this gateway. |+| [OpenHands/software-agent-sdk#5922](https://github.com/OpenHands/software-agent-sdk/issues/5922) | mapped | reproduced | [F12.activate](../feature-map/F12-model-router.md) | Two real Model Router Set active attempts over owned corrupt settings showed an error toast; activation returned 500 with settings-load/path detail instead of409. Corrupt bytes were not overwritten, original bytes restored in finally, doctor healthy. No model call. |+| [OpenHands/software-agent-sdk#5914](https://github.com/OpenHands/software-agent-sdk/issues/5914) | mapped | blocked | [F20.app-backend-view](../feature-map/F20-canvas-apps.md) | Canvas app-backend sessions forward an owned app backend HTTP response. Duplicate Date/Server response headers require an admitted bridge session and a reviewed backend emitting those headers; default demo without the bridge and a direct backend fetch do not prove the proxy contract. A Linux Agent Server is required; native macOS artifact backends are unsupported. |+| [OpenHands/software-agent-sdk#5910](https://github.com/OpenHands/software-agent-sdk/issues/5910) | out-of-scope | out-of-scope | — | The reported malformed artifact prepare409-versus500 path has no current Canvas UI producer: CanvasExtensionsService offers install/enable/update/uninstall and bridge sessions but no backend/prepare operation. The backend admin REST contract belongs to the SDK; generic arbitrary app API access does not establish a mapped builtin consumer. |+| [OpenHands/software-agent-sdk#5899](https://github.com/OpenHands/software-agent-sdk/issues/5899) | mapped | blocked | [F13.openhands-options](../feature-map/F13-agent-profiles.md) | Profile tool selection consumes the Agent Server tool catalog usability, including browser_tool_set. A genuine custom-only browser installation is required; standard user Chrome on this Mac would mask this condition. |+| [OpenHands/software-agent-sdk#5897](https://github.com/OpenHands/software-agent-sdk/issues/5897) | mapped | reproduced | [F19.refresh](../feature-map/F19-plugins.md) | Two real plugin Update clicks after moving only the owned source aside surfaced error toasts; refresh returned 500 instead of handled400. Installed version/enabled state stayed unchanged. Source restored, own plugin uninstalled and fixture removed; no model call. |+| [OpenHands/software-agent-sdk#5894](https://github.com/OpenHands/software-agent-sdk/issues/5894) | out-of-scope | out-of-scope | — | Canvas SkillsService only lists user/project skills plus bundled catalog. Add skill is an instructions-only /add-skill composer flow; Canvas does not call POST/api/skills/install. A malformed SDK management-install body is not a current direct Canvas consumer. The bundled add-skill skill instructs a GitHub fetch script writing workspace/.agents/skills; it does not use SDK management install. |+| [OpenHands/software-agent-sdk#5890](https://github.com/OpenHands/software-agent-sdk/issues/5890) | mapped | blocked | [F10.edit / F10.default-used](../feature-map/F10-llm-profiles.md) | Local conversation profile selection now fetches fresh profile config then calls switchLLM, so the managed refresh-hook loss after that switch remains a consumer. Refused-loopback errors cannot establish a real401 refresh/retry outcome. |+| [OpenHands/software-agent-sdk#5889](https://github.com/OpenHands/software-agent-sdk/issues/5889) | mapped | blocked | [F05.attach-files](../feature-map/F05-composer.md) | Same-named Canvas attachment upload overwrites existing workspace file; a failed mid-stream upload can destroy previous contents. |+| [OpenHands/software-agent-sdk#5883](https://github.com/OpenHands/software-agent-sdk/issues/5883) | out-of-scope | out-of-scope | — | Canvas SecretsService calls SettingsClient.listSecrets() without agent_profile_id. Profile secret selection lists global names and saves secret_refs; the failing scoped secret-list query is not emitted. |+| [OpenHands/software-agent-sdk#5882](https://github.com/OpenHands/software-agent-sdk/issues/5882) | out-of-scope | out-of-scope | [F10.model-picker / F10.add-models](../feature-map/F10-llm-profiles.md) | API retains Bedrock-only deepseek.* IDs, but current Canvas filters provider options to provider/ names: all 3 affected IDs were absent from real Basic and Bulk add. SDK enumeration bug is not a current Canvas picker failure. |+| [OpenHands/software-agent-sdk#5881](https://github.com/OpenHands/software-agent-sdk/issues/5881) | out-of-scope | out-of-scope | — | Canvas MetaProfileEditor refuses Save unless classifier_model is nonempty and prompt_template contains{{instance_text}}; handleSave always sends classes:[] plus that template. It cannot produce the classless/template-less management body in this issue. |+| [OpenHands/software-agent-sdk#5880](https://github.com/OpenHands/software-agent-sdk/issues/5880) | mapped | reproduced | [F10.subscription](../feature-map/F10-llm-profiles.md) | Two real Connect clicks with the backend routed through an owned refused-loopback proxy returned 500 rather than502/504, showing the friendly subscription error plus All connection attempts failed. UV_OFFLINE used the already-cached runtime. Each EXIT restored the original environment and restarted; doctor passed. No subscription account or model call; genuine pending-token poll remains untested. |+| [OpenHands/software-agent-sdk#5879](https://github.com/OpenHands/software-agent-sdk/issues/5879) | mapped | reproduced | [F12.activate](../feature-map/F12-model-router.md) | With actual global ACP settings, two real Set active clicks each toasted activation success while API/settings active_meta_profile stayed null. Original owned settings restored byte-for-byte; no ACP subprocess or model call. |+| [OpenHands/software-agent-sdk#5878](https://github.com/OpenHands/software-agent-sdk/issues/5878) | mapped | blocked | [F12.create-first / F12.list](../feature-map/F12-model-router.md) | Router save/list uses the SDK MetaProfileStore, whose path helper mishandles a leading~ OH_PERSISTENCE_DIR. Default Canvas launcher uses an absolute owned path, so that normal run cannot exercise the report. |+| [OpenHands/software-agent-sdk#5873](https://github.com/OpenHands/software-agent-sdk/issues/5873) | mapped | reproduced | [F13.create / F13.edit / F13.set-active](../feature-map/F13-agent-profiles.md) | Real Canvas create/edit accepted QA_SUFFIX.json; two concurrency-only overwrites each minted a different UUID at revision0. On SDK 1.55 the dangling active pointer repaired to default, so no persistent dangling-pointer claim. No model call. |+| [OpenHands/software-agent-sdk#5872](https://github.com/OpenHands/software-agent-sdk/issues/5872) | out-of-scope | out-of-scope | — | This issue is the exported TypeScript type mirror: missing/extra declared fields and materialize diagnostics. Canvas does not call materialize or read resolved_mcp_servers; its editor preserves the real stored object through mergeAgentProfileSaveInput and sends disabled_skills-compatible fields. No concrete current rendered failure is supplied. |+| [OpenHands/software-agent-sdk#5871](https://github.com/OpenHands/software-agent-sdk/issues/5871) | out-of-scope | out-of-scope | — | SDK internal _seed_default_llm_profile docstring promises a nonexistent materialize field. The report explicitly says runtime is fine; Canvas neither shows that docstring nor consumes materialize diagnostics. |+| [OpenHands/software-agent-sdk#5870](https://github.com/OpenHands/software-agent-sdk/issues/5870) | out-of-scope | out-of-scope | — | Canvas BashService.searchEvents sends kind,command_id,sort_order,page_id,limit, never timestamp__gte/lt. Its event-history timestamp filters target conversation events, not BashEventService. The report's non-UTC Bash filter branch has no Canvas consumer. |+| [OpenHands/software-agent-sdk#5863](https://github.com/OpenHands/software-agent-sdk/issues/5863) | out-of-scope | out-of-scope | — | Canvas OAuth UI starts a probe, opens its authorization URL and polls status. It does not call the public submitOAuthCallback wrapper anywhere; malformed port URLs in that unused REST callback POST have no current UI entry point. |+| [OpenHands/software-agent-sdk#5862](https://github.com/OpenHands/software-agent-sdk/issues/5862) | mapped | blocked | [F10.create-validation](../feature-map/F10-llm-profiles.md) | Profile Save displays provider preflight error messages; its actual validation request uses this SDK exception mapping. A refused endpoint does not generate the real DeepSeek401 shape and cannot establish the error classification. |+| [OpenHands/software-agent-sdk#5861](https://github.com/OpenHands/software-agent-sdk/issues/5861) | out-of-scope | out-of-scope | — | Canvas profile rows use names returned by the list and send those canonical names to Delete/Rename. The report bypasses FK guards by manually adding.json to the target URL; the UI has no alias-URL input and does not emit that alternate spelling. |+| [OpenHands/software-agent-sdk#5860](https://github.com/OpenHands/software-agent-sdk/issues/5860) | out-of-scope | out-of-scope | — | The double-submit/replay endpoint is not called by current Canvas OAuth UI. authorizeOAuth opens the provider popup and polls status; submitOAuthCallback has only its service definition/test callers. No current UI callback submission can be retried. |+| [OpenHands/software-agent-sdk#5849](https://github.com/OpenHands/software-agent-sdk/issues/5849) | mapped | blocked | [F03.folder-browser / F03.manage-workspaces](../feature-map/F03-home.md) | Canvas WorkspacesService add/remove workspace and parent calls back Home and new-conversation pickers. Read-only registry storage can turn a duplicate re-add or stale deletion into 500 instead of idempotent 200/404. File inode/mtime changes alone are not UI proof. Requires an owned read-only registry plus real duplicate/stale picker action; surface owns note, this ledger owns row. |+| [OpenHands/software-agent-sdk#5847](https://github.com/OpenHands/software-agent-sdk/issues/5847) | out-of-scope | out-of-scope | — | Canvas activates Agent/Model Router profiles through dedicated activate endpoints with known list IDs/names, rather than PATCHing arbitrary active_agent_profile_id/active_meta_profile pointers. The malformed unknown-pointer PATCH is not emitted by current settings saves. |+| [OpenHands/software-agent-sdk#5846](https://github.com/OpenHands/software-agent-sdk/issues/5846) | out-of-scope | out-of-scope | — | OpenAPI docstrings advertise400 instead of422 for invalid secret names; runtime is already422. Canvas does not display these descriptions or branch on the documented400, and its form validates names. This is SDK documentation-only. |+| [OpenHands/software-agent-sdk#5845](https://github.com/OpenHands/software-agent-sdk/issues/5845) | mapped | blocked | [F08.vscode-link](../feature-map/F08-workspace-files-and-changes.md) | Local VS Code controls call VSCodeClient.getUrl with the raw conversation workspace directory; reserved query characters can select the wrong folder. Requires a genuinely running editor exposed by this origin and an owned workspace containing &, #, ?, %, + or space. Cloud exposed_urls is a separate producer. Surface owns the note, this ledger owns the row. |+| [OpenHands/software-agent-sdk#5844](https://github.com/OpenHands/software-agent-sdk/issues/5844) | out-of-scope | out-of-scope | — | The reported401 is specific to the slashless Docker workspace root. Canvas useWorkspaceSession receives a trailing-slash base from RemoteWorkspace.startWorkspaceSession; joinWorkspaceUrl preserves it for empty and file paths. It never generates the affected slashless root. |+| [OpenHands/software-agent-sdk#5843](https://github.com/OpenHands/software-agent-sdk/issues/5843) | mapped | reproduced | [F08.uncommitted-changes / F08.diff-view-modes](../feature-map/F08-workspace-files-and-changes.md) | Canvas Commits in repo/src showed mixed paths README.md,new.py,src/app.py,src/util.py. Expanding src/app.py requested repo/src/src/app.py and returned 400 twice, so the real file diff could not load. Root repo control showed five correctly based paths. |+| [OpenHands/software-agent-sdk#5841](https://github.com/OpenHands/software-agent-sdk/issues/5841) | out-of-scope | out-of-scope | [F08.diff-deleted-file](../feature-map/F08-workspace-files-and-changes.md) | Actual Canvas Commits lists deleted src/util.py; expanding it displays the defined deleted-file placeholder and sends zero /git/diff requests. Source intentionally disables working-tree D diff. SDK endpoint400 is bypassed by this consumer; no claim server fixed. |+| [OpenHands/software-agent-sdk#5839](https://github.com/OpenHands/software-agent-sdk/issues/5839) | out-of-scope | out-of-scope | — | Current Local Canvas already fetches fresh config and invokes switchLLM with a fresh profile:<name>:<uuidv4()> usage_id, deliberately avoiding the SDK switch_profile registry cache. Cloud uses its app-server route, a different transport from this Agent Server reproducer. |+| [OpenHands/software-agent-sdk#5838](https://github.com/OpenHands/software-agent-sdk/issues/5838) | mapped | blocked | [F10.default-used](../feature-map/F10-llm-profiles.md) | Canvas live profile selection calls switchLLM on the running conversation and current source does not force persisted readback. The GET/base_state disagreement during a real running tool step remains relevant. |+| [OpenHands/software-agent-sdk#5837](https://github.com/OpenHands/software-agent-sdk/issues/5837) | mapped | blocked | [F03.folder-browser](../feature-map/F03-home.md) | Workspace folder picker follows search_subdirs cursor; case-only directory twins can repeat cursor and hide later choices. |+| [OpenHands/software-agent-sdk#5836](https://github.com/OpenHands/software-agent-sdk/issues/5836) | out-of-scope | out-of-scope | — | Canvas awaits upload success but ignores file_size, building uploaded/skipped names from File inputs. TS LocalConversation write_file is not Canvas agent runtime. |+| [OpenHands/software-agent-sdk#5835](https://github.com/OpenHands/software-agent-sdk/issues/5835) | mapped | blocked | [F27.overview-identity](../feature-map/F27-workspace-tools.md) | Local Overview Git reads use a conversation-scoped BashClient.executeCommand with a10s timeout. With a short configured idle TTL and an overlapping eviction pass, scoped work can be killed without final output. Recheck a genuine still-running Canvas Git probe on that owned runtime; agent Terminal events are a different producer from the standalone async sleep in the report. |+| [OpenHands/software-agent-sdk#5834](https://github.com/OpenHands/software-agent-sdk/issues/5834) | out-of-scope | out-of-scope | — | Explicitly an OpenAPI/generated description bug; Canvas does not display that endpoint prose. Runtime behavior is intentionally unchanged. |+| [OpenHands/software-agent-sdk#5833](https://github.com/OpenHands/software-agent-sdk/issues/5833) | out-of-scope | out-of-scope | — | The 500 branch requires a malformed non-UUID conversation id. Canvas gets its scoped Bash id from a real active conversation and route guards resolve invalid routes before useLocalGitInfo enables; the builtin producer does not issue this malformed-id call. SDK parameter validation is independent. |+| [OpenHands/software-agent-sdk#5825](https://github.com/OpenHands/software-agent-sdk/issues/5825) | out-of-scope | out-of-scope | — | Canvas does not issue PUT credential-bindings/CODEX_AUTH_JSON to regular unloaded conversations. Native secret delivery uses structured conversation secrets; ACP binding lifecycle is server/Cloud infrastructure and the report depends on an externally injected non-ACP binding. |+| [OpenHands/software-agent-sdk#5824](https://github.com/OpenHands/software-agent-sdk/issues/5824) | mapped | blocked | [F05.slash-goal](../feature-map/F05-composer.md) | Canvas exposes /goal --max and iteration/max banner; failed judge then Resume can show2/1 and overrun cap. |+| [OpenHands/software-agent-sdk#5823](https://github.com/OpenHands/software-agent-sdk/issues/5823) | out-of-scope | out-of-scope | — | Current ACP composer picker lists only availableAcpModels from the live session/discovery and has no free-text unknown-model switch input. The reported request manually sends an unoffered ID to the API. |+| [OpenHands/software-agent-sdk#5822](https://github.com/OpenHands/software-agent-sdk/issues/5822) | out-of-scope | out-of-scope | — | Canvas never calls agent_final_response nor server navigate. Message editing forks instead, so reported abandoned-branch route has no consumer. |+| [OpenHands/software-agent-sdk#5820](https://github.com/OpenHands/software-agent-sdk/issues/5820) | mapped | blocked | [F05.context-window-meter](../feature-map/F05-composer.md) | Compact context is exposed without checking server condenser support; null/no-op condenser can return500. |+| [OpenHands/software-agent-sdk#5813](https://github.com/OpenHands/software-agent-sdk/issues/5813) | out-of-scope | out-of-scope | [F05.slash-btw](../feature-map/F05-composer.md) | Actual /btw submission on an owned conversation deleted out of band made a200 batch metadata preflight returning missing conversation, then zero ask_agent requests. fetchConversationData throws V1 conversation not found before the affected route. After metadata polling Canvas returned to list. No assertion that SDK unknown-ID endpoint is fixed. |+| [OpenHands/software-agent-sdk#5810](https://github.com/OpenHands/software-agent-sdk/issues/5810) | out-of-scope | out-of-scope | — | No current Canvas caller emits /load_plugin. Plugins are attached during conversation creation and read in metadata; the issue requires calling the dynamic load_plugin API twice. Creation-time plugin merging is a different path and not established by this report. |+| [OpenHands/software-agent-sdk#5809](https://github.com/OpenHands/software-agent-sdk/issues/5809) | out-of-scope | out-of-scope | — | Canvas already tracks highest contiguous sequence, not highest seen. For0,2,1 its cursor stays0 until1 arrives, preventing reported reconnect skip; no claim wire ordering fixed. |+| [OpenHands/software-agent-sdk#5808](https://github.com/OpenHands/software-agent-sdk/issues/5808) | out-of-scope | out-of-scope | — | SDK TypeScript docstrings incorrectly promise409 pre-first-run; Canvas adapter already documents200 deferred behavior and does not rely on those stale comments. Runtime is explicitly unchanged. |+| [OpenHands/software-agent-sdk#5807](https://github.com/OpenHands/software-agent-sdk/issues/5807) | out-of-scope | out-of-scope | — | ACP composer switches only from concrete model entries returned by its session/discovery (or the explicit Agent default path); it has no blank/free-text model switch input. The reported whitespace-only REST body is not emitted by this current consumer; settings profile creation is a different API path. |+| [OpenHands/software-agent-sdk#5806](https://github.com/OpenHands/software-agent-sdk/issues/5806) | mapped | blocked | [F06.branch-from-here](../feature-map/F06-agent-activity.md) | User-message edit/fork reads parent with getEvent; displayed event whose file disappears can fail500 before forking. Canvas does not use batch event getter. |+| [OpenHands/software-agent-sdk#5804](https://github.com/OpenHands/software-agent-sdk/issues/5804) | out-of-scope | out-of-scope | — | Supported Canvas send paths always send user role; no UI posts assistant/system/tool role through SendMessageRequest. |+| [OpenHands/software-agent-sdk#5801](https://github.com/OpenHands/software-agent-sdk/issues/5801) | out-of-scope | out-of-scope | — | Reported loss requires timestamp__gte continuation or legacy resend_mode=since. Current reconnect uses session after_seq and unfiltered count; timestampLt older paging is a distinct contract. |+| [OpenHands/software-agent-sdk#5800](https://github.com/OpenHands/software-agent-sdk/issues/5800) | mapped | reproduced | [F07.delete-confirm / F07.rename-inline](../feature-map/F07-conversation-page.md) | Actual Delete confirmation after out-of-band removal issuedDELETE400 and showed Bad Request. Stale inline rename control PATCH200 was followed by missing metadata and explicit not-found UI, so Canvas does not falsely confirm that rename. |+| [OpenHands/software-agent-sdk#5799](https://github.com/OpenHands/software-agent-sdk/issues/5799) | out-of-scope | out-of-scope | — | Canvas trims title before PATCH, so whitespace UI input sends empty string and takes422, rather than SDK whitespace200. Existing Canvas 17061 owns blank-rename UI behavior. |+| [OpenHands/software-agent-sdk#5791](https://github.com/OpenHands/software-agent-sdk/issues/5791) | out-of-scope | out-of-scope | — | Canvas constructs structured secret source objects for conversation creation and sends string values only to the update endpoint that intentionally accepts them. Number/null/boolean malformed values and plain-string create bodies are not current UI payloads. |+| [OpenHands/software-agent-sdk#5790](https://github.com/OpenHands/software-agent-sdk/issues/5790) | mapped | reproduced | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | Real99-child parent load issued200 batch and opened planner plus parent session sockets. Adding the100th genuine idle child then full reload issued500 batch, displayed Internal Server Error and opened only parent socket, losing discovered planner. No model runs; no browser storage read. |+| [OpenHands/software-agent-sdk#5789](https://github.com/OpenHands/software-agent-sdk/issues/5789) | mapped | blocked | [F25.add-agent-server / F25.health-status](../feature-map/F25-backends-and-cloud.md) | Deferred-init server CORS origins delivered via/api/init are ignored by middleware; a real non-localhost Canvas origin then cannot admit/probe that backend. Requires that owned warm-pool server and actual non-localhost browser origin; the default loopback frontend is always allowed. |+| [OpenHands/software-agent-sdk#5784](https://github.com/OpenHands/software-agent-sdk/issues/5784) | mapped | blocked | [F08.files-rich-plain / F08.files-fallbacks / F08.files-open-new-window](../feature-map/F08-workspace-files-and-changes.md) | Path-prefixed local backend workspace previews use RemoteWorkspace.startWorkspaceSession and cookie authentication; the cookie path omits root_path and prefixed iframe/img/file requests return 401. Needs an authenticated prefixed server, real workspace asset and browser cookie readback. Surface owns the note, this ledger owns the row. |+| [OpenHands/software-agent-sdk#5740](https://github.com/OpenHands/software-agent-sdk/issues/5740) | mapped | blocked | [F06.error-events](../feature-map/F06-agent-activity.md) | Genuine Local report records SDK ConversationErrorEvent when Responses parser lacks item_id; current1.55 accesses event.item_id in display-delta processing. |+| [OpenHands/software-agent-sdk#5736](https://github.com/OpenHands/software-agent-sdk/issues/5736) | out-of-scope | out-of-scope | — | Canvas does not call the TypeScript getHooksForEvent matching helper. Hook execution is server-side Python; Canvas only renders HookExecutionEvent matchers. The TS regex helper mismatch has no current executing Canvas consumer. |+| [OpenHands/software-agent-sdk#5731](https://github.com/OpenHands/software-agent-sdk/issues/5731) | mapped | blocked | [F26.cli-public-needs-key](../feature-map/F26-runtime-variants.md) | Canvas launcher forwards nonempty LOCAL_BACKEND_API_KEY without trimming; Agent Server config retains whitespace-only session keys. Recheck an owned loopback-only public deployment with a whitespace key, actual authentication and a normal-key control. The default generated strong key does not exercise it; SDK cipher fallback is distinct when Canvas supplies OH_SECRET_KEY. |+| [OpenHands/software-agent-sdk#5729](https://github.com/OpenHands/software-agent-sdk/issues/5729) | mapped | blocked | [F22.setup-repository-list](../feature-map/F22-automation-creation.md) | Automation 1.20 prompt/plugin preset workers call RemoteWorkspace.clone_repos, which reads lowercase provider names. An uppercase-only documented secret can leave an owned private repo unauthenticated. Needs a disposable authorized repository or reviewed token-gated local Git fixture through the real preset worker, plus a refused-loopback model configuration if initialization is needed; no paid inference is authorized. |+| [OpenHands/software-agent-sdk#5718](https://github.com/OpenHands/software-agent-sdk/issues/5718) | mapped | historical-reproduction | [F19.enabled-autoload](../feature-map/F19-plugins.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5711](https://github.com/OpenHands/software-agent-sdk/issues/5711) | out-of-scope | out-of-scope | — | Canvas does not use TypeScript RemoteState; it owns event/state hydration through its frontend stores. |+| [OpenHands/software-agent-sdk#5652](https://github.com/OpenHands/software-agent-sdk/issues/5652) | out-of-scope | out-of-scope | — | Python RemoteConversation cached getter behavior; Canvas consumes typed HTTP/WebSocket payloads rather than Python RemoteState. |+| [OpenHands/software-agent-sdk#5651](https://github.com/OpenHands/software-agent-sdk/issues/5651) | out-of-scope | out-of-scope | — | Node 22 ConversationEventStream reconnect path; Canvas uses browser sockets and requires Node >=24 for its launcher. |+| [OpenHands/software-agent-sdk#5650](https://github.com/OpenHands/software-agent-sdk/issues/5650) | out-of-scope | out-of-scope | — | RemoteWorkspace.downloadAsBlob/fileDownload are not Canvas download consumers; Canvas uses binary-safe FileClient.downloadFile. |+| [OpenHands/software-agent-sdk#5649](https://github.com/OpenHands/software-agent-sdk/issues/5649) | out-of-scope | out-of-scope | — | Canvas builds its create payload and calls ConversationClient directly, not TypeScript ConversationManager.createConversation. |+| [OpenHands/software-agent-sdk#5648](https://github.com/OpenHands/software-agent-sdk/issues/5648) | mapped | blocked | [F20.app-backend-view](../feature-map/F20-canvas-apps.md) | Actual server_info on pinned Agent Server1.54.0 has no canvas_app_backend_bridge_v1 and null app_backend_ingress_url; reviewed demo displays App backend view unavailable. Need an admitted bridge-capable server, platform artifact and bridge-enabled deployment to drive this backend subprotocol/health/orphan report through Canvas. |+| [OpenHands/software-agent-sdk#5647](https://github.com/OpenHands/software-agent-sdk/issues/5647) | out-of-scope | out-of-scope | — | OpenAI-compatible gateway per-turn usage contract; Canvas does not consume /v1/chat/completions gateway usage. |+| [OpenHands/software-agent-sdk#5645](https://github.com/OpenHands/software-agent-sdk/issues/5645) | mapped | historical-reproduction | [F19.add-plugin](../feature-map/F19-plugins.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5644](https://github.com/OpenHands/software-agent-sdk/issues/5644) | out-of-scope | out-of-scope | — | Canvas's LlmSettingsLocalView.handleSave explicitly skips ProfilesService.validateProfile when connectionId is set. The linked-draft validation request in this SDK report is not emitted by the mapped Canvas profile-save path. |+| [OpenHands/software-agent-sdk#5643](https://github.com/OpenHands/software-agent-sdk/issues/5643) | mapped | historical-reproduction | [F17.custom-add / F17.agent-uses-server](../feature-map/F17-mcp-servers.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5642](https://github.com/OpenHands/software-agent-sdk/issues/5642) | mapped | historical-reproduction | [F08.uncommitted-changes](../feature-map/F08-workspace-files-and-changes.md) | Tracked my notes.md modification makes scoped git/changes return400 and hides changes. |+| [OpenHands/software-agent-sdk#5639](https://github.com/OpenHands/software-agent-sdk/issues/5639) | mapped | blocked | [F05.llm-profile-picker](../feature-map/F05-composer.md) | A working model, a profile switch before the first run, and nonzero usage/cost after that turn. |+| [OpenHands/software-agent-sdk#5638](https://github.com/OpenHands/software-agent-sdk/issues/5638) | mapped | historical-reproduction | [F08.files-tree](../feature-map/F08-workspace-files-and-changes.md) | A real1800-file listing over1MiB drops README.md and creates a partial-path bb folder before the2000-file cap. |+| [OpenHands/software-agent-sdk#5636](https://github.com/OpenHands/software-agent-sdk/issues/5636) | out-of-scope | out-of-scope | — | Canvas direct profile switches assign a new profile:<name>:<uuid> usage_id; the raw switch_llm default/already-registered usage_id collision is not its request shape. |+| [OpenHands/software-agent-sdk#5635](https://github.com/OpenHands/software-agent-sdk/issues/5635) | mapped | historical-reproduction | [F06.reconnect](../feature-map/F06-agent-activity.md) | prepare-for-sandbox-pause returned 204 but the original communicating session socket stayed open for over10s. A newly opened Canvas tab received a new run:false message; the old tab did not. DELETE returned 200 and GET404, but both held sockets stayed open over10s. Canvas later redirected through metadata polling; no claim of permanently stale navigation is made. |+| [OpenHands/software-agent-sdk#5633](https://github.com/OpenHands/software-agent-sdk/issues/5633) | mapped | historical-reproduction | [F26.remote-backend](../feature-map/F26-runtime-variants.md) | Temporarily replaced only the owned Agent Server with the same pinned interpreter and state under genuine default Rich logging (LOG_JSON/CI/LOG_RICH_TRACEBACKS unset). Canvas Send uses tools=[] and a dummy LLM pointing at closed 127.0.0.1:9; 60 concurrent direct /alive probes had maximum latency 2.091 s while the conversation ended error. No provider reachable. Restored normal launcher and doctor afterward. |+| [OpenHands/software-agent-sdk#5632](https://github.com/OpenHands/software-agent-sdk/issues/5632) | mapped | historical-reproduction | [F11.list](../feature-map/F11-provider-connections.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5631](https://github.com/OpenHands/software-agent-sdk/issues/5631) | out-of-scope | out-of-scope | — | SDK event-search short-kind alias contract; current Canvas event history requests do not send that short kind filter. |+| [OpenHands/software-agent-sdk#5629](https://github.com/OpenHands/software-agent-sdk/issues/5629) | mapped | historical-reproduction | [F07.rename-inline](../feature-map/F07-conversation-page.md) | B was started with shared storage before A renamed an idle conversation from BEFORE_B to AFTER_B. A PATCH and disk confirmed AFTER_B. After own A graceful stop, native B events/count returned 200 and rewrote disk title to its BEFORE_B startup copy. Real Add Backend→B card showed BEFORE_B in Canvas. The native first-load request is explicit; an earlier sandbox-pause variant did not reach a valid first load and is not the claimed reproduction. |+| [OpenHands/software-agent-sdk#5628](https://github.com/OpenHands/software-agent-sdk/issues/5628) | mapped | historical-reproduction | [F19.enabled-autoload](../feature-map/F19-plugins.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5627](https://github.com/OpenHands/software-agent-sdk/issues/5627) | mapped | blocked | [F05.slash-goal](../feature-map/F05-composer.md) | A working model goal loop active during graceful backend restart, then goal-status and resume readback. |+| [OpenHands/software-agent-sdk#5626](https://github.com/OpenHands/software-agent-sdk/issues/5626) | not-reproduced | not-reproduced | [F06.stop-resume](../feature-map/F06-agent-activity.md) | During a genuine held dummy-loopback model request, Canvas Stop immediately persisted paused in REST and base_state.json. After SIGKILL of the verified own Agent Server listener and an owned restart, REST/disk and the reloaded Stopped UI retained paused. The bounded accept-only endpoint read no bytes and generated no model reply. |+| [OpenHands/software-agent-sdk#5624](https://github.com/OpenHands/software-agent-sdk/issues/5624) | mapped | historical-reproduction | [F11.create-validation / F11.edit](../feature-map/F11-provider-connections.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5622](https://github.com/OpenHands/software-agent-sdk/issues/5622) | mapped | historical-reproduction | [F14.create / F14.edit / F14.delete](../feature-map/F14-secrets.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5607](https://github.com/OpenHands/software-agent-sdk/issues/5607) | mapped | historical-reproduction | [F17.credential-probe](../feature-map/F17-mcp-servers.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5590](https://github.com/OpenHands/software-agent-sdk/issues/5590) | out-of-scope | out-of-scope | — | Python create_app(Config) lifespan embedding contract; Canvas launches agent-server using its CLI/environment configuration. |+| [OpenHands/software-agent-sdk#5539](https://github.com/OpenHands/software-agent-sdk/issues/5539) | mapped | historical-reproduction | [F07.download-zip](../feature-map/F07-conversation-page.md) | A20-second three-client real loopback overlapping receiver made929 attempts:31 completed corrupt200 ZIPs,106HTTP500,269ingress502,133RemoteProtocolError,388 validZIPs and 2 intentionally deadline-aborted reads. Both ambiguous Content-Length differences were excluded from the public claim. Actual Canvas Download during overlap produced a ZIP with inspectable entries; that particular UI download is not claimed corrupt. A separate sequential2000-request pass returned validZIPs throughout. |+| [OpenHands/software-agent-sdk#5533](https://github.com/OpenHands/software-agent-sdk/issues/5533) | mapped | blocked | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | A model key for real agent-driven first-command observations is unavailable. |+| [OpenHands/software-agent-sdk#5523](https://github.com/OpenHands/software-agent-sdk/issues/5523) | mapped | blocked | [F05.image-only-send](../feature-map/F05-composer.md) | A working DeepSeek V4.1 Flash vision profile and another fallback profile, with an image-bearing turn. |+| [OpenHands/software-agent-sdk#5501](https://github.com/OpenHands/software-agent-sdk/issues/5501) | mapped | not-run | [F20.rail-label](../feature-map/F20-canvas-apps.md) | Existing Agent Server nav_label omission note and F20.rail-label recipe preserved. |+| [OpenHands/software-agent-sdk#5498](https://github.com/OpenHands/software-agent-sdk/issues/5498) | mapped | not-run | [F11.delete-stale-reference](../feature-map/F11-provider-connections.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5497](https://github.com/OpenHands/software-agent-sdk/issues/5497) | mapped | blocked | [F11.rotate](../feature-map/F11-provider-connections.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5496](https://github.com/OpenHands/software-agent-sdk/issues/5496) | mapped | historical-reproduction | [F19.refresh / F19.start-conversation / F19.enabled-autoload](../feature-map/F19-plugins.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5495](https://github.com/OpenHands/software-agent-sdk/issues/5495) | mapped | not-run | [F15.condenser-kind](../feature-map/F15-agent-behavior-settings.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5492](https://github.com/OpenHands/software-agent-sdk/issues/5492) | mapped | blocked | [F05.slash-goal](../feature-map/F05-composer.md) | A real goal judge response containing complete as the string false rather than a JSON boolean. |+| [OpenHands/software-agent-sdk#5475](https://github.com/OpenHands/software-agent-sdk/issues/5475) | out-of-scope | out-of-scope | — | Enterprise SDK dependency-bump CI permissions; no Canvas feature or user flow. |+| [OpenHands/software-agent-sdk#5466](https://github.com/OpenHands/software-agent-sdk/issues/5466) | mapped | historical-reproduction | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5465](https://github.com/OpenHands/software-agent-sdk/issues/5465) | mapped | blocked | [F06.branch-from-here](../feature-map/F06-agent-activity.md) | A working cipher-backed model conversation, then fork and a successful turn on the fork. |+| [OpenHands/software-agent-sdk#5447](https://github.com/OpenHands/software-agent-sdk/issues/5447) | mapped | blocked | [F06.error-events](../feature-map/F06-agent-activity.md) | Anthropic Claude Opus5.5 credentials and a real trailing-assistant-prefill request. |+| [OpenHands/software-agent-sdk#5375](https://github.com/OpenHands/software-agent-sdk/issues/5375) | mapped | blocked | [F17.install-oauth / F17.agent-uses-server](../feature-map/F17-mcp-servers.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5373](https://github.com/OpenHands/software-agent-sdk/issues/5373) | mapped | blocked | [F26.remote-backend](../feature-map/F26-runtime-variants.md) | A separately hosted default-Rich backend and the original cyclic-exception peer/backlog driver are unavailable. |+| [OpenHands/software-agent-sdk#5354](https://github.com/OpenHands/software-agent-sdk/issues/5354) | mapped | blocked | [F06.event-groups](../feature-map/F06-agent-activity.md) | A live model turn mixing valid, invalid and valid tool calls in one response. |+| [OpenHands/software-agent-sdk#5330](https://github.com/OpenHands/software-agent-sdk/issues/5330) | mapped | blocked | [F05.slash-btw](../feature-map/F05-composer.md) | A working caching provider and comparable ordinary-message versus /btw prompt/cache records. |+| [OpenHands/software-agent-sdk#5322](https://github.com/OpenHands/software-agent-sdk/issues/5322) | mapped | blocked | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | A model key for genuine terminal-tool execution is unavailable; the PowerShell half additionally requires Windows. |+| [OpenHands/software-agent-sdk#5310](https://github.com/OpenHands/software-agent-sdk/issues/5310) | mapped | blocked | [F26.docker-conversation-runtime](../feature-map/F26-runtime-variants.md) | Docker daemon/per-conversation image and real auxiliary-title/ACP credentials are unavailable. |+| [OpenHands/software-agent-sdk#5307](https://github.com/OpenHands/software-agent-sdk/issues/5307) | mapped | blocked | [F26.docker-conversation-runtime](../feature-map/F26-runtime-variants.md) | Docker daemon/per-conversation image is unavailable; a host/container git fixture is required. |+| [OpenHands/software-agent-sdk#5279](https://github.com/OpenHands/software-agent-sdk/issues/5279) | mapped | blocked | [F23.run-linked-conversation](../feature-map/F23-automation-detail.md) | An authenticated Cloud automation workspace/callback context and owned stable conversation are unavailable. |+| [OpenHands/software-agent-sdk#5252](https://github.com/OpenHands/software-agent-sdk/issues/5252) | out-of-scope | out-of-scope | — | Cross-version test-fixture gap; no reported user-visible failure to reproduce. |+| [OpenHands/software-agent-sdk#5225](https://github.com/OpenHands/software-agent-sdk/issues/5225) | mapped | blocked | [F06.error-events](../feature-map/F06-agent-activity.md) | A real provider-rejected poisoned image tool_result in a disposable persisted history, followed by retry. |+| [OpenHands/software-agent-sdk#5207](https://github.com/OpenHands/software-agent-sdk/issues/5207) | mapped | blocked | [F11.agent-uses-connection / F11.rotate](../feature-map/F11-provider-connections.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5181](https://github.com/OpenHands/software-agent-sdk/issues/5181) | mapped | blocked | [F10.schema-validation](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5167](https://github.com/OpenHands/software-agent-sdk/issues/5167) | mapped | blocked | [F13.acp-conversation](../feature-map/F13-agent-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5152](https://github.com/OpenHands/software-agent-sdk/issues/5152) | not-reproduced | not-reproduced | [F27.browser-screenshot](../feature-map/F27-workspace-tools.md) | Using the actual owned Agent Server 1.55.0 interpreter and private HOME, BrowserToolExecutor() constructs successfully. The resolved browser-use 0.11.13, mcp 1.30.0 and fastmcp 3.4.8 do not hit the reported mcp 2.x import failure. This is the SDK report’s explicit constructor reproduction; it does not establish a successful Canvas browser operation. |+| [OpenHands/software-agent-sdk#5138](https://github.com/OpenHands/software-agent-sdk/issues/5138) | out-of-scope | out-of-scope | — | Retired TypeScript cloud-proxy body API; current Canvas makes direct Cloud App-API calls and uses typed runtime clients. |+| [OpenHands/software-agent-sdk#5108](https://github.com/OpenHands/software-agent-sdk/issues/5108) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | An Agent Server configured with the native Gemini edit tool set and a real tool edit of an invalid-UTF8 fixture; Gemini ACP uses a different external executor. |+| [OpenHands/software-agent-sdk#5100](https://github.com/OpenHands/software-agent-sdk/issues/5100) | mapped | historical-reproduction | [F10.create-validation](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#5096](https://github.com/OpenHands/software-agent-sdk/issues/5096) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model invoking canonical file-editor str_replace/insert through a real symlink, then undo. |+| [OpenHands/software-agent-sdk#5094](https://github.com/OpenHands/software-agent-sdk/issues/5094) | mapped | blocked | [F07.open-by-url](../feature-map/F07-conversation-page.md) | An authenticated Codex ACP session whose original session/load fails while persisted history remains. |+| [OpenHands/software-agent-sdk#5092](https://github.com/OpenHands/software-agent-sdk/issues/5092) | mapped | blocked | [F06.confirmation-mode](../feature-map/F06-agent-activity.md) | A working model proposing a real confirmation-gated marker action, then a revocation user message. |+| [OpenHands/software-agent-sdk#5074](https://github.com/OpenHands/software-agent-sdk/issues/5074) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A native delegated child still running during an owned server crash and parent recovery. |+| [OpenHands/software-agent-sdk#5038](https://github.com/OpenHands/software-agent-sdk/issues/5038) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A live canonical file-editor view/edit on the specific UTF8 sample alignment; Canvas Files preview uses a separate NUL sniff, not this validator. |+| [OpenHands/software-agent-sdk#5027](https://github.com/OpenHands/software-agent-sdk/issues/5027) | out-of-scope | out-of-scope | — | PR linked-issue readiness-check repository identity; repository CI tooling. |+| [OpenHands/software-agent-sdk#4990](https://github.com/OpenHands/software-agent-sdk/issues/4990) | mapped | blocked | [F06.stop-resume](../feature-map/F06-agent-activity.md) | An authenticated Claude Code ACP turn executing separate marker-writing tools while Pause is pressed. |+| [OpenHands/software-agent-sdk#4987](https://github.com/OpenHands/software-agent-sdk/issues/4987) | mapped | blocked | [F04.card](../feature-map/F04-conversation-list.md) | A working autotitle model and captured title-generation input/output; Canvas only consumes the resulting server title. |+| [OpenHands/software-agent-sdk#4983](https://github.com/OpenHands/software-agent-sdk/issues/4983) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A runtime without ripgrep and a live agent grep call using extended regex and escaped literal punctuation. |+| [OpenHands/software-agent-sdk#4965](https://github.com/OpenHands/software-agent-sdk/issues/4965) | mapped | blocked | [F06.error-events](../feature-map/F06-agent-activity.md) | A strict OpenAI-compatible provider and a real history containing blank text or empty content arrays. |+| [OpenHands/software-agent-sdk#4963](https://github.com/OpenHands/software-agent-sdk/issues/4963) | mapped | blocked | [F27.terminal-output](../feature-map/F27-workspace-tools.md) | Native Windows PowerShell terminal runtime is unavailable. |+| [OpenHands/software-agent-sdk#4934](https://github.com/OpenHands/software-agent-sdk/issues/4934) | mapped | blocked | [F10.schema-validation / F10.default-used](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#4912](https://github.com/OpenHands/software-agent-sdk/issues/4912) | out-of-scope | out-of-scope | — | SDK CI path-gating and downstream test selection; no Canvas runtime feature. |+| [OpenHands/software-agent-sdk#4901](https://github.com/OpenHands/software-agent-sdk/issues/4901) | mapped | blocked | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | An owned authenticated ACP provider runtime/command chain is unavailable. |+| [OpenHands/software-agent-sdk#4893](https://github.com/OpenHands/software-agent-sdk/issues/4893) | mapped | blocked | [F07.status-menu](../feature-map/F07-conversation-page.md) | A disposable sandbox/VM frozen mid-run with a live model and persisted ownership state. |+| [OpenHands/software-agent-sdk#4891](https://github.com/OpenHands/software-agent-sdk/issues/4891) | not-reproduced | not-reproduced | [F17.agent-uses-server](../feature-map/F17-mcp-servers.md) | Two real initializations with a missing stdio executable and refused HTTP sibling retained the healthy qa_echo tool, persisted user messages and reached the expected loopback model error. Successful provider replies remain outside this check. |+| [OpenHands/software-agent-sdk#4847](https://github.com/OpenHands/software-agent-sdk/issues/4847) | out-of-scope | out-of-scope | — | Python RemoteConversation.fork() wrapper title parsing; Canvas calls the typed REST fork endpoint directly. |+| [OpenHands/software-agent-sdk#4846](https://github.com/OpenHands/software-agent-sdk/issues/4846) | out-of-scope | out-of-scope | — | Node ESM WebSocket constructor selection; Canvas browser selects window.WebSocket. |+| [OpenHands/software-agent-sdk#4811](https://github.com/OpenHands/software-agent-sdk/issues/4811) | out-of-scope | out-of-scope | — | The report is about standalone SDK fork title injection into ConversationState.tags; it explicitly separates Agent Server StoredConversation title handling used by Canvas. |+| [OpenHands/software-agent-sdk#4810](https://github.com/OpenHands/software-agent-sdk/issues/4810) | out-of-scope | out-of-scope | — | Canvas sets policy/analyzer/secrets at creation or through account settings/resume secrets; it does not call these per-conversation mutation endpoints. |+| [OpenHands/software-agent-sdk#4800](https://github.com/OpenHands/software-agent-sdk/issues/4800) | mapped | historical-reproduction | [F07.rename-inline](../feature-map/F07-conversation-page.md) | After a successful baseline UI rename, chmod0500 on only the owned conversation directory made the next actual UI rename fail with PermissionError/HTTP 500. GET nevertheless reported the rejected title while meta.json retained the baseline. Permissions were restored; graceful restart then persisted the rejected in-memory title. Tags/autotitle branches were not tested. |+| [OpenHands/software-agent-sdk#4777](https://github.com/OpenHands/software-agent-sdk/issues/4777) | mapped | blocked | [F06.stop-resume](../feature-map/F06-agent-activity.md) | A live batch with a second real tool waiting on a declared-resource lock when Stop is pressed. |+| [OpenHands/software-agent-sdk#4776](https://github.com/OpenHands/software-agent-sdk/issues/4776) | out-of-scope | out-of-scope | — | There is no structuredContent, structured_content, result_meta or resultMeta consumer under current src/. The report demonstrates SDK conversion/persistence metadata loss, without a current mapped Canvas renderer or API consumer for those fields. |+| [OpenHands/software-agent-sdk#4774](https://github.com/OpenHands/software-agent-sdk/issues/4774) | out-of-scope | out-of-scope | — | Python MessageEvent.visualize() CLI terminal rendering; Canvas renders MessageEvent content in its own React components. |+| [OpenHands/software-agent-sdk#4765](https://github.com/OpenHands/software-agent-sdk/issues/4765) | out-of-scope | out-of-scope | — | The affected redact_text_secrets helper is used by SDK command and ACP diagnostic logging. This report supplies only a direct Python helper call and no affected Canvas-rendered event, endpoint field or user control; F14 secret CRUD does not call it. |+| [OpenHands/software-agent-sdk#4731](https://github.com/OpenHands/software-agent-sdk/issues/4731) | out-of-scope | out-of-scope | — | TypeScript package bundler compatibility for require(ws); current Canvas browser event transport is separate and the frozen Canvas production build succeeds. |+| [OpenHands/software-agent-sdk#4709](https://github.com/OpenHands/software-agent-sdk/issues/4709) | not-reproduced | not-reproduced | [F15.agent-context-datetime](../feature-map/F15-agent-behavior-settings.md) | Two UI launches with an explicit2001 settings timestamp produced fresh runtime datetimes and no2001 timestamp in events. Two natural-stale controls also used fresh datetimes. The original setting was restored. |+| [OpenHands/software-agent-sdk#4705](https://github.com/OpenHands/software-agent-sdk/issues/4705) | mapped | blocked | [F23.run-status-polling / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Affected client/server versions, GitHub tool-definition event fixture and a model key for genuine run/status comparison are unavailable. |+| [OpenHands/software-agent-sdk#4670](https://github.com/OpenHands/software-agent-sdk/issues/4670) | out-of-scope | out-of-scope | — | Python WebSocketCallbackClient thread shutdown; not the Canvas browser WebSocket path. |+| [OpenHands/software-agent-sdk#4668](https://github.com/OpenHands/software-agent-sdk/issues/4668) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A live planner create/edit yielding PlanningFileEditorObservation with old/new contents for the chat diff. |+| [OpenHands/software-agent-sdk#4667](https://github.com/OpenHands/software-agent-sdk/issues/4667) | out-of-scope | out-of-scope | — | Tom conversation index/checkpoint implementation; no mapped Canvas feature or supported Canvas agent path identified. |+| [OpenHands/software-agent-sdk#4666](https://github.com/OpenHands/software-agent-sdk/issues/4666) | out-of-scope | out-of-scope | — | The report concerns the backend browser-use rrweb recording controller/flush-task lifecycle. Canvas has no backend recording/replay surface or duplicate-start action; control-openhands browser record is a separate verification-browser implementation and is not this SDK consumer. |+| [OpenHands/software-agent-sdk#4665](https://github.com/OpenHands/software-agent-sdk/issues/4665) | out-of-scope | out-of-scope | — | Canvas SkillsService.getSkills uses the bundled @openhands/extensions catalog and the local /api/skills list with load_public:false. It does not call the SDK skills marketplace loader/endpoint whose default.json assumption is reported. |+| [OpenHands/software-agent-sdk#4664](https://github.com/OpenHands/software-agent-sdk/issues/4664) | out-of-scope | out-of-scope | — | OpenHandsCloudWorkspace is a Python workspace object; Canvas uses TypeScript clients and Cloud runtime API transport, not that object or its cached Python connection fields. |+| [OpenHands/software-agent-sdk#4663](https://github.com/OpenHands/software-agent-sdk/issues/4663) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | Concurrent real glob/tool calls in distinct workspaces with the Python fallback active. |+| [OpenHands/software-agent-sdk#4661](https://github.com/OpenHands/software-agent-sdk/issues/4661) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | Two native task-enabled conversations with same-name project subagent definitions and a working model for delegation. |+| [OpenHands/software-agent-sdk#4660](https://github.com/OpenHands/software-agent-sdk/issues/4660) | out-of-scope | out-of-scope | — | Canvas Model Router uses meta-profiles and buildRouterAtStartSystemSuffix to guide profile selection; it does not instantiate the Python RouterLLM or call its completion/acompletion methods. The reported SDK class path is not its implementation. |+| [OpenHands/software-agent-sdk#4659](https://github.com/OpenHands/software-agent-sdk/issues/4659) | out-of-scope | out-of-scope | — | Canvas BashService.listOutputs reads every result page sorted by timestamp, and no Canvas consumer polls BashOutput with order__gt. The reported duplicate order values affect reconstruction clients using that cursor; the current Canvas logs consumer does not use the affected order contract. |+| [OpenHands/software-agent-sdk#4658](https://github.com/OpenHands/software-agent-sdk/issues/4658) | out-of-scope | out-of-scope | — | The report is transactionality/resource unwind in POST /api/init for a deferred warm-pool coordinator. Canvas neither calls that init endpoint nor owns its staged runtime publication, and no Canvas-observable failure is given in the report; this is a coordinator/Agent Server lifecycle contract before the UI attaches. |+| [OpenHands/software-agent-sdk#4629](https://github.com/OpenHands/software-agent-sdk/issues/4629) | mapped | historical-reproduction | [F13.acp-credentials](../feature-map/F13-agent-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#4613](https://github.com/OpenHands/software-agent-sdk/issues/4613) | mapped | blocked | [F10.custom-model](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#4583](https://github.com/OpenHands/software-agent-sdk/issues/4583) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | A working model invoking file-editor insert at EOF on a file with no trailing newline. |+| [OpenHands/software-agent-sdk#4575](https://github.com/OpenHands/software-agent-sdk/issues/4575) | mapped | blocked | [F26.runtime-services](../feature-map/F26-runtime-variants.md) | A model key and controlled bounded workload/resource envelope for a real Canvas terminal run are unavailable. |+| [OpenHands/software-agent-sdk#4562](https://github.com/OpenHands/software-agent-sdk/issues/4562) | mapped | blocked | [F08.files-load-error](../feature-map/F08-workspace-files-and-changes.md) | A plain-HTTP non-loopback same-origin browser deployment with real workspace-cookie observation. |+| [OpenHands/software-agent-sdk#4558](https://github.com/OpenHands/software-agent-sdk/issues/4558) | mapped | blocked | [F06.error-events](../feature-map/F06-agent-activity.md) | A working encrypted saved fallback profile and real primary-model failure that triggers fallback resolution. |+| [OpenHands/software-agent-sdk#4540](https://github.com/OpenHands/software-agent-sdk/issues/4540) | mapped | blocked | [F06.tool-visualizers](../feature-map/F06-agent-activity.md) | Synthetic provider credentials and a GLM/Kimi turn emitting the raw tool_call wrapper. |+| [OpenHands/software-agent-sdk#4537](https://github.com/OpenHands/software-agent-sdk/issues/4537) | mapped | blocked | [F04.card](../feature-map/F04-conversation-list.md) | A working model with native TaskTool delegation holding the parent state lock during list reload. |+| [OpenHands/software-agent-sdk#4533](https://github.com/OpenHands/software-agent-sdk/issues/4533) | not-reproduced | not-reproduced | [F13.active-drives-conversation](../feature-map/F13-agent-profiles.md) | Four actual seeded-default UI launches retained the dummy LLM key in runtime state and reached only refused-loopback model errors. This verifies propagation, not successful provider authentication. |+| [OpenHands/software-agent-sdk#4532](https://github.com/OpenHands/software-agent-sdk/issues/4532) | mapped | blocked | [F06.confirmation-mode](../feature-map/F06-agent-activity.md) | A real confirmation-paused action persisted across backend restart, then approval and a working provider call. |+| [OpenHands/software-agent-sdk#4511](https://github.com/OpenHands/software-agent-sdk/issues/4511) | mapped | blocked | [F10.schema-validation](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#4498](https://github.com/OpenHands/software-agent-sdk/issues/4498) | mapped | blocked | [F05.context-window-meter](../feature-map/F05-composer.md) | A real condensation splitting an action/result pair, followed by the next provider request. |+| [OpenHands/software-agent-sdk#4388](https://github.com/OpenHands/software-agent-sdk/issues/4388) | mapped | blocked | [F23.run-status-polling / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Frontend BashService does not send order__gt, but automation 1.20 launch preparation and prompt/plugin preset clones use (Async)RemoteWorkspace.execute_command. Its remote_workspace_mixin output generator sends order__gt after received chunks. A bounded genuine worker/preparation or Git clone crossing the 100-event output boundary must be correlated with final BashOutput and Canvas run status; frontend Logs-only paging and a one-output command do not cover it. |+| [OpenHands/software-agent-sdk#4382](https://github.com/OpenHands/software-agent-sdk/issues/4382) | mapped | blocked | [F05.context-window-meter](../feature-map/F05-composer.md) | An authenticated ACP provider emitting cache/thought token usage or reset/stalled cumulative cost updates. |+| [OpenHands/software-agent-sdk#4363](https://github.com/OpenHands/software-agent-sdk/issues/4363) | mapped | historical-reproduction | [F19.refresh](../feature-map/F19-plugins.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#4260](https://github.com/OpenHands/software-agent-sdk/issues/4260) | mapped | blocked | [F23.run-linked-conversation / F23.run-logs-modal](../feature-map/F23-automation-detail.md) | Owned Cloud reviewer/webhook deployment, authorized disposable review target and model credentials are unavailable. |+| [OpenHands/software-agent-sdk#4252](https://github.com/OpenHands/software-agent-sdk/issues/4252) | not-reproduced | not-reproduced | [F18.personal-skill-dirs](../feature-map/F18-skills.md) | A persistence-folder skill appeared in the real Skills page and in initialized conversation events on four UI launches; no successful model reply or historical Docker mount behavior is claimed. |+| [OpenHands/software-agent-sdk#4250](https://github.com/OpenHands/software-agent-sdk/issues/4250) | mapped | blocked | [F10.custom-model / F10.default-used](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#4247](https://github.com/OpenHands/software-agent-sdk/issues/4247) | mapped | blocked | [F10.custom-model](../feature-map/F10-llm-profiles.md) | Preserve the Oct 12 observation and its exact limits as history; SDK pin changed1.54→1.55, so this row is not a current reproduction or current not-reproduced verdict. |+| [OpenHands/software-agent-sdk#4246](https://github.com/OpenHands/software-agent-sdk/issues/4246) | not-reproduced | not-reproduced | [F17.agent-uses-server](../feature-map/F17-mcp-servers.md) | Two actual90s stdio hang fixtures left initialization after the real MCP timeout, persisted system/user events and reached visible loopback model error. This settles the current Local initialization loop only; successful replies and historical Cloud/SSE remain unverified. |+| [OpenHands/software-agent-sdk#4245](https://github.com/OpenHands/software-agent-sdk/issues/4245) | out-of-scope | out-of-scope | — | Legacy docker.openhands.dev/openhands/openhands v0 webhook deployment, absent from current Agent Canvas stack. |+| [OpenHands/software-agent-sdk#4158](https://github.com/OpenHands/software-agent-sdk/issues/4158) | out-of-scope | out-of-scope | — | Current ACP composer offers ACP model choices and not an OpenHands LLM-profile switcher; it cannot request the cross-agent-kind switch_profile described by the old1.4.0 report. Local OpenHands profile switching now uses fresh switchLLM configs, while ACP remains a separate picker. |+| [OpenHands/software-agent-sdk#4080](https://github.com/OpenHands/software-agent-sdk/issues/4080) | mapped | blocked | [F04.card](../feature-map/F04-conversation-list.md) | A copied persisted conversation with its original unregistered observation kind and missing tool-module registration. |+| [OpenHands/software-agent-sdk#4077](https://github.com/OpenHands/software-agent-sdk/issues/4077) | mapped | blocked | [F06.reload-mid-run](../feature-map/F06-agent-activity.md) | A working stream interrupted/retried or delivered to a controlled stalled subscriber, with delta/state-order capture. |++## Final checks++- OpenHands/OpenHands#18324: recorded. Real main build/assets/root-CEvL3aWX.css contains .bg-black{background-color:var(--color-black)}, no .bg-opacity-50 utility, and supported slash-opacity utilities. Current routed settings uses SettingsMobileHub.+- OpenHands/software-agent-sdk#5820 (F05.context-window-meter): blocked. Actual idle Usage panel has cost0 but usage=null, so no context card/Compact context action. A known gpt-4o dummy model with max_input_tokens128000 and genuine refused-loopback send still yields no context control. Separate direct API diagnostic returned 500 No condenser configured; this is not claimed as UI reproduction.+- OpenHands/software-agent-sdk#5790 (F06.tool-visualizers): fail. Real99-child parent load issued200 batch and opened planner plus parent session sockets. Adding the100th genuine idle child then full reload issued500 batch, displayed Internal Server Error and opened only parent socket, losing discovered planner. No model runs; no browser storage read.+- OpenHands/software-agent-sdk#5841 (F08.diff-deleted-file): pass. Actual Canvas Commits lists deleted src/util.py; expanding it displays the defined deleted-file placeholder and sends zero /git/diff requests. Source intentionally disables working-tree D diff. SDK endpoint400 is bypassed by this consumer; no claim server fixed.+- OpenHands/software-agent-sdk#5843 (F08.uncommitted-changes): fail. Canvas Commits in repo/src showed mixed paths README.md,new.py,src/app.py,src/util.py. Expanding src/app.py requested repo/src/src/app.py and returned 400 twice, so the real file diff could not load. Root repo control showed five correctly based paths.+- OpenHands/software-agent-sdk#5641 (F08.uncommitted-changes): fail. On pinned 1.55.0, actual Commits tab for verified plain parent workspace shows No commits yet; scoped /git/changes returned 200 while owned child proj has modified README and untracked notes. Closed upstream fix is not in current pin.+- OpenHands/software-agent-sdk#5813 (F05.slash-btw): pass. Actual /btw submission on an owned conversation deleted out of band made a200 batch metadata preflight returning missing conversation, then zero ask_agent requests. fetchConversationData throws V1 conversation not found before the affected route. After metadata polling Canvas returned to list. No assertion that SDK unknown-ID endpoint is fixed.+- OpenHands/software-agent-sdk#5800 (F07.delete-confirm): fail. Actual Delete confirmation after out-of-band removal issuedDELETE400 and showed Bad Request. Stale inline rename control PATCH200 was followed by missing metadata and explicit not-found UI, so Canvas does not falsely confirm that rename.+- OpenHands/OpenHands#18421 (F05.attach-files): fail. Actual Home New Worktree send posted attachment to original repository path and returned 422 path must be inside conversation workspace; created runtime workspace is /tmp/conversation-worktrees/<id>/qa-surface-attachment. Same file sent inside that existing conversation uploaded to worktree path200. Only dummy refused-loopback profile used.+- OpenHands/software-agent-sdk#5630 (F03.launch-in-workspace): pass. Original owned project subagent references missing skill. Actual Home Local Repo launch on1.55 returnedPOST/conversations201, fetched metadata200 and opened conversation route; former missing-skill500 did not occur. Later dummy-model error is separate and expected. Pinned _register_definition catches ValueError, warns and skips invalid agent.+- OpenHands/OpenHands#17955 (F10.create-prefill): pass. UI showed0.4 then provider-linked Save persisted config.temperature 0.4 and num_retries0. No validation/model call.+- OpenHands/OpenHands#18204 (F11.row-menu-focus): fail. Two real Chrome opens kept activeElement=testid provider-connection-menu-trigger; first ArrowDown kept same focus.+- OpenHands/software-agent-sdk#5882 (F10.model-picker): pass. API retains Bedrock-only deepseek.* IDs, but current Canvas filters provider options to provider/ names: all 3 affected IDs were absent from real Basic and Bulk add. SDK enumeration bug is not a current Canvas picker failure.+- OpenHands/software-agent-sdk#5873 (F13.create): fail. Real Canvas create/edit accepted QA_SUFFIX.json; two concurrency-only overwrites each minted a different UUID at revision0. On SDK 1.55 the dangling active pointer repaired to default, so no persistent dangling-pointer claim. No model call.+- OpenHands/software-agent-sdk#5879 (F12.activate): fail. With actual global ACP settings, two real Set active clicks each toasted activation success while API/settings active_meta_profile stayed null. Original owned settings restored byte-for-byte; no ACP subprocess or model call.+- OpenHands/software-agent-sdk#5922 (F12.activate): fail. Two real Model Router Set active attempts over owned corrupt settings showed an error toast; activation returned 500 with settings-load/path detail instead of409. Corrupt bytes were not overwritten, original bytes restored in finally, doctor healthy. No model call.+- OpenHands/software-agent-sdk#5897 (F19.refresh): fail. Two real plugin Update clicks after moving only the owned source aside surfaced error toasts; refresh returned 500 instead of handled400. Installed version/enabled state stayed unchanged. Source restored, own plugin uninstalled and fixture removed; no model call.+- F13.llm-pill-precedence: pass. Current1.55 runtime confirms per-launch LLM override preserves named profile and active deletion restores default.+- OpenHands/software-agent-sdk#5880 (F10.subscription): fail. Two real Connect clicks with the backend routed through an owned refused-loopback proxy returned 500 rather than502/504, showing the friendly subscription error plus All connection attempts failed. UV_OFFLINE used the already-cached runtime. Each EXIT restored the original environment and restarted; doctor passed. No subscription account or model call; genuine pending-token poll remains untested.+- check: pass. ok;28 map files;778 stable IDs;no problems+- check: pass. ok;28 map files;778 stable IDs;no problems+- check: pass. ok;28 map files;778 stable IDs;no problems+- check: pass. ok;28 map files;778 stable IDs;no problems+- check: pass. ok;28 map files;778 stable IDs;no problems+- check: pass. ok;28 map files;778 stable IDs;no problems+- check: pass. ok;28 map files;778 stable IDs;no problems+- check: pass. ok;28 map files;778 stable IDs;no problems+- check: pass. ok;28 map files;778 stable IDs;no problems+- OpenHands/OpenHands#17564 (F21.phone): pass. Heading 347×28 and 277×28; no page overflow.+- OpenHands/OpenHands#17128 (F21.kebab-menu): pass. Phone menu x196..356,y510..746; desktop x1248..1408,y666..902. Delete stays inside each.+- OpenHands/OpenHands#16155 (F21.list-row): pass. Run now is icon-only 32×32 px; no document overflow.+- OpenHands/OpenHands#15833 (F21.add-menu): pass. New Automation form and assistance input visible.+- OpenHands/software-agent-sdk#5152 (F27.browser-screenshot): pass. Exit 0; BrowserToolExecutor; SDK/Agent Server 1.55.0, browser-use 0.11.13, mcp 1.30.0, fastmcp 3.4.8.+- OpenHands/OpenHands#17563 (F23.phone): pass. Name/Save/Close all in viewport, dialog count 0, document overflow 0; inner schedule-control overflow is outside this old header-clipping check.+- OpenHands/OpenHands#18196 (F20.enable-confirm): pass. Focus remains in the dialog in both directions; Escape/Cancel restore disabled Enable switch.+- OpenHands/OpenHands#18461 (F28.schedule-presets): fail. Custom shows 0 9 * * *.+- OpenHands/OpenHands#18481 (F28.error-contrast): pass. Minimum 4.623:1 on Solarized Light; every theme exceeds4.5:1.+- OpenHands/OpenHands#16217 (F26.runtime-services): pass. Automation health 200 and doctor passes on fixture PATH and restored PATH.+- OpenHands/OpenHands#17898: pass. Two real opt-in Save/reload cycles on the hard-DNT build produced zero PostHog/telemetry-host requests and no PostHog storage keys. The second reloaded form showed consent checked; consent was then restored off. This covers the hard VITE_DO_NOT_TRACK build, not a browser-only DNT preference.+- OpenHands/software-agent-sdk#4533: pass. Four actual seeded-default UI launches retained the dummy LLM key in runtime state and reached only refused-loopback model errors. This verifies propagation, not successful provider authentication.+- OpenHands/software-agent-sdk#4252: pass. A persistence-folder skill appeared in the real Skills page and in initialized conversation events on four UI launches; no successful model reply or historical Docker mount behavior is claimed.+- OpenHands/software-agent-sdk#4709: pass. Two UI launches with an explicit2001 settings timestamp produced fresh runtime datetimes and no2001 timestamp in events. Two natural-stale controls also used fresh datetimes. The original setting was restored.+- OpenHands/OpenHands#15694: pass. The owned skill stayed disabled through two eight-second samples and reloads (four aria-checked=false readbacks). This bounded sample does not settle every intermittent timing.+- OpenHands/software-agent-sdk#4891: pass. Two real initializations with a missing stdio executable and refused HTTP sibling retained the healthy qa_echo tool, persisted user messages and reached the expected loopback model error. Successful provider replies remain outside this check.+- OpenHands/software-agent-sdk#4246: pass. Two actual90s stdio hang fixtures left initialization after the real MCP timeout, persisted system/user events and reached visible loopback model error. This settles the current Local initialization loop only; successful replies and historical Cloud/SSE remain unverified.+- OpenHands/OpenHands#16578: pass. Two genuine Canvas Bedrock starts reached the missing named-profile failure instead of a missing boto3 import. AWS metadata was disabled and both runtime/base endpoints refused loopback. This covers current Local1.55 named-profile import, not AWS credentials, successful inference or Linux fresh installs.+- OpenHands/software-agent-sdk#5882: pass. API retains Bedrock-only deepseek.* IDs, but current Canvas filters provider options to provider/ names: all 3 affected IDs were absent from real Basic and Bulk add. SDK enumeration bug is not a current Canvas picker failure.+- OpenHands/software-agent-sdk#5873: fail. Real Canvas create/edit accepted QA_SUFFIX.json; two concurrency-only overwrites each minted a different UUID at revision0. On SDK 1.55 the dangling active pointer repaired to default, so no persistent dangling-pointer claim. No model call.+- OpenHands/software-agent-sdk#5879: fail. With actual global ACP settings, two real Set active clicks each toasted activation success while API/settings active_meta_profile stayed null. Original owned settings restored byte-for-byte; no ACP subprocess or model call.+- OpenHands/software-agent-sdk#5922: fail. Two real Model Router Set active attempts over owned corrupt settings showed an error toast; activation returned 500 with settings-load/path detail instead of409. Corrupt bytes were not overwritten, original bytes restored in finally, doctor healthy. No model call.+- OpenHands/software-agent-sdk#5897: fail. Two real plugin Update clicks after moving only the owned source aside surfaced error toasts; refresh returned 500 instead of handled400. Installed version/enabled state stayed unchanged. Source restored, own plugin uninstalled and fixture removed; no model call.+- OpenHands/OpenHands#17619: pass. Two current local trials retained the same open page across an owned backend restart. Each session advanced its after_seq cursor to the prior REST event count minus one, with an observed old-socket close, a new communicating connection and one visible recovery marker. Metadata only: no claim about payload replay, provider streams or Cloud.+- OpenHands/software-agent-sdk#5626: pass. During a genuine held dummy-loopback model request, Canvas Stop immediately persisted paused in REST and base_state.json. After SIGKILL of the verified own Agent Server listener and an owned restart, REST/disk and the reloaded Stopped UI retained paused. The bounded accept-only endpoint read no bytes and generated no model reply.+- OpenHands/software-agent-sdk#5880: fail. Two real Connect clicks with the backend routed through an owned refused-loopback proxy returned 500 rather than502/504, showing the friendly subscription error plus All connection attempts failed. UV_OFFLINE used the already-cached runtime. Each EXIT restored the original environment and restarted; doctor passed. No subscription account or model call; genuine pending-token poll remains untested.+- Feature-map structure: pass. 28 families and 778 stable IDs; no problems.+- Route and feature coverage: pass. 31 routes and 29 feature component directories accounted for; context-menu is an explicit shared-component exclusion.+- Cited test IDs: pass. All 1103 cited IDs resolve.+- Harness unit tests: pass. 64/64 pass under Node 24.19.0.+- Bounded Python fixture tests: pass. 8/8 pass with the installed SDK Python, including shared-storage and trajectory-overlap fixtures.+- Owned-run teardown: pass. Both task launchers stopped; every owned port closed; private fixtures purged;117 evidence files retained.+- Final main application build: pass. npm run build passed at 5385a76fcc9d7ded5cc5fbcdc62b7e464a9136c8, whose product tree matches main d9ecec24181a0dd7b83b5d87561916abb2fe0e8c. Node 24.19.0.+- Final main library build: pass. npm run build:lib passed at the same merged revision.+- Final theme and scrim regression tests: pass. 27/27 tests pass in color-themes, settings-navigation and context-meter at the final merged product revision. This checks source/build contracts, not fresh visual fix proof.+- PR scope and ancestry: pass. All differences from final main are under .agents/skills/verify-openhands/. Both main and the maintainer PR merge are ancestors; original October 12 records remain byte-for-byte unchanged.++Every frozen-inventory issue has one supported accounting outcome. Dates and verification limits distinguish historical notes from fresh observations.diff --git a/.agents/skills/verify-openhands/references/backfills/evidence/16948-after-refetch.png b/.agents/skills/verify-openhands/references/backfills/evidence/16948-after-refetch.pngnew file mode 100644index 0000000000..a0dd3cda68Binary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/16948-after-refetch.png differdiff --git a/.agents/skills/verify-openhands/references/backfills/evidence/16948-before-refetch.png b/.agents/skills/verify-openhands/references/backfills/evidence/16948-before-refetch.pngnew file mode 100644index 0000000000..b3d141e70eBinary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/16948-before-refetch.png differdiff --git a/.agents/skills/verify-openhands/references/backfills/evidence/17619-reconnected.png b/.agents/skills/verify-openhands/references/backfills/evidence/17619-reconnected.pngnew file mode 100644index 0000000000..9a0b5ec4c1Binary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/17619-reconnected.png differdiff --git a/.agents/skills/verify-openhands/references/backfills/evidence/17619-socket-cursor.json b/.agents/skills/verify-openhands/references/backfills/evidence/17619-socket-cursor.jsonnew file mode 100644index 0000000000..52f4538d57--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/evidence/17619-socket-cursor.json@@ -0,0 +1,139 @@+{+ "target": "fa1f724ab5737a17453ad6081129eb91c5a423c6",+ "sdk": "1.54.0",+ "events_count_before_restart": 3,+ "latest_sequence_before_restart": 2,+ "page_reloaded": false,+ "recovery_marker_dom_count": 1,+ "events": [+ {+ "ts": "2026-10-12T09:28:58.489Z",+ "kind": "websocket-created",+ "appOrigin": true,+ "connectionId": "42b4ed0f95dd5c07",+ "afterSeq": "-1",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:28:58.490Z",+ "kind": "websocket-open",+ "appOrigin": true,+ "connectionId": "42b4ed0f95dd5c07",+ "afterSeq": "-1",+ "receivedFrames": 0,+ "sentFrames": 1,+ "observed": "first-frame"+ },+ {+ "ts": "2026-10-12T09:29:08.789Z",+ "kind": "websocket-close",+ "appOrigin": true,+ "connectionId": "42b4ed0f95dd5c07",+ "afterSeq": "-1",+ "receivedFrames": 5,+ "sentFrames": 1+ },+ {+ "ts": "2026-10-12T09:29:09.801Z",+ "kind": "websocket-created",+ "appOrigin": true,+ "connectionId": "f563e2f6058190ba",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:09.801Z",+ "kind": "websocket-error",+ "appOrigin": true,+ "connectionId": "f563e2f6058190ba",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:09.805Z",+ "kind": "websocket-close",+ "appOrigin": true,+ "connectionId": "f563e2f6058190ba",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:12.179Z",+ "kind": "websocket-created",+ "appOrigin": true,+ "connectionId": "a1a6fcd9f74e114d",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:12.180Z",+ "kind": "websocket-error",+ "appOrigin": true,+ "connectionId": "a1a6fcd9f74e114d",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:12.183Z",+ "kind": "websocket-close",+ "appOrigin": true,+ "connectionId": "a1a6fcd9f74e114d",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:17.106Z",+ "kind": "websocket-created",+ "appOrigin": true,+ "connectionId": "e16f1230363f6eb7",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:17.106Z",+ "kind": "websocket-error",+ "appOrigin": true,+ "connectionId": "e16f1230363f6eb7",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:17.111Z",+ "kind": "websocket-close",+ "appOrigin": true,+ "connectionId": "e16f1230363f6eb7",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:26.434Z",+ "kind": "websocket-created",+ "appOrigin": true,+ "connectionId": "9a07f9bd0056c27d",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 0+ },+ {+ "ts": "2026-10-12T09:29:26.465Z",+ "kind": "websocket-open",+ "appOrigin": true,+ "connectionId": "9a07f9bd0056c27d",+ "afterSeq": "2",+ "receivedFrames": 0,+ "sentFrames": 1,+ "observed": "first-frame"+ }+ ],+ "privacy": "Page and socket session URLs are omitted. Only the numeric reconnect cursor, lifecycle timestamps and frame counts are published."+}diff --git a/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-pause-durability.json b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-pause-durability.jsonnew file mode 100644index 0000000000..943d7db0d8--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-pause-durability.json@@ -0,0 +1,15 @@+{+ "before_crash": {+ "api": "paused",+ "disk": "paused"+ },+ "after_restart": {+ "api": "paused",+ "disk": "paused"+ },+ "agent_server_listener_identity_verified": true,+ "ancestry_includes_owned_launcher": true,+ "accepted_connections": 1,+ "received_bytes_recorded": 0,+ "ui_after_restart": "Stopped"+}diff --git a/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-100.png b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-100.pngnew file mode 100644index 0000000000..831cc80e0eBinary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-100.png differdiff --git a/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-99.png b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-99.pngnew file mode 100644index 0000000000..4bdb8efe03Binary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-99.png differdiff --git a/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-boundary.json b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-boundary.jsonnew file mode 100644index 0000000000..b962ec5f1c--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-planner-boundary.json@@ -0,0 +1,30 @@+{+ "checked_at": "2026-10-16",+ "product_revision": "b34ac59e691d16354c14291325d5bd44b37226c5",+ "runtime_revision": "054521cbff8f676d442f8471983cb2ae7361db2d",+ "run_id": "backfill-18334-refresh",+ "pins": {+ "agentServer": "1.55.0",+ "automation": "1.20.0",+ "canvas": "1.27.1"+ },+ "issue": "https://github.com/OpenHands/software-agent-sdk/issues/5790",+ "arrangement": "Genuine idle child conversations, one tagged as the planner; no model turn or fabricated browser hints. Full page reload at each boundary.",+ "control": {+ "children": 99,+ "batch_http_status": 200,+ "observed_session_sockets": [+ "parent",+ "planner"+ ]+ },+ "test": {+ "children": 100,+ "batch_http_status": 500,+ "visible_error": "Internal Server Error",+ "new_observed_session_sockets": [+ "parent"+ ]+ },+ "limits": "Socket metadata only, no frame bodies. Screenshots and this readback are a current observation, not a baseline/fix proof."+}diff --git a/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-reconnect-readback.json b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-reconnect-readback.jsonnew file mode 100644index 0000000000..855954fde7--- /dev/null+++ b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-reconnect-readback.json@@ -0,0 +1,18 @@+[+ {+ "trial": 1,+ "events_before_restart": 7,+ "initial_after_seq": -1,+ "reconnected_after_seq": 6,+ "recovery_message_visible_count": 1,+ "observed_close_and_new_communicating_connection": true+ },+ {+ "trial": 2,+ "events_before_restart": 3,+ "initial_after_seq": -1,+ "reconnected_after_seq": 2,+ "recovery_message_visible_count": 1,+ "observed_close_and_new_communicating_connection": true+ }+]diff --git a/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-schedule-custom.png b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-schedule-custom.pngnew file mode 100644index 0000000000..d6f03027e8Binary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-schedule-custom.png differdiff --git a/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-settings-keyboard.png b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-settings-keyboard.pngnew file mode 100644index 0000000000..5737a0be03Binary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/2026-10-16-settings-keyboard.png differdiff --git a/.agents/skills/verify-openhands/references/backfills/evidence/network-state-and-refetch.gif b/.agents/skills/verify-openhands/references/backfills/evidence/network-state-and-refetch.gifnew file mode 100644index 0000000000..fc47bd4741Binary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/network-state-and-refetch.gif differdiff --git a/.agents/skills/verify-openhands/references/backfills/evidence/network-state-and-refetch.mp4 b/.agents/skills/verify-openhands/references/backfills/evidence/network-state-and-refetch.mp4new file mode 100644index 0000000000..3fbb96e7d0Binary files /dev/null and b/.agents/skills/verify-openhands/references/backfills/evidence/network-state-and-refetch.mp4 differdiff --git a/.agents/skills/verify-openhands/references/bug-backfill.md b/.agents/skills/verify-openhands/references/bug-backfill.mdnew file mode 100644index 0000000000..f84d30828a--- /dev/null+++ b/.agents/skills/verify-openhands/references/bug-backfill.md@@ -0,0 +1,125 @@+# Accounting for existing bugs++[OpenHands/OpenHands#18309](https://github.com/OpenHands/OpenHands/issues/18309)+backfills bugs that predate the map's maintenance baseline. Freeze the current+`main` SHA and the open `bug` issue inventory separately: a Git revision fixes+the product under test; the inventory timestamp fixes which issues were open.+Do not move the maintenance baseline during this backfill.++The inventory includes every open Canvas `bug`, excluding pull requests, and+SDK or automation bugs with a concrete mapped Canvas consumer. Read the report+and the relevant source before assigning a family. Bugs in a removed legacy+product, CI, documentation, unused SDK client paths or build/repository tooling+without a current Canvas consumer are outside this map, with an explicit+reason. User-visible installation and startup failures belong to the runtime+map even when their cause is packaging or build output. Missing credentials,+OS or fixture data are verification prerequisites, not reasons to exclude a+Canvas bug.++`ready-for-dev` means the issue is sufficiently specified and accepted for+implementation. Use it to prioritize live checks. It was introduced recently,+so older unmarked reports with usable reproduction steps are also candidates.+Do not add or remove readiness, priority, `bug` or the existing `reproduced`+label as a side effect of map bookkeeping.++## Three outcomes++| Outcome | Required record | GitHub label | Color |+| --- | --- | --- | --- |+| Mapped | Existing note, a live `Known failure` beside the unchanged expected result, or a `Known issue` Gotchas entry naming the concrete prerequisite when no recipe can exercise it. | `feature-map:mapped` | Blue `#1D76DB` |+| Not reproduced | A live run on the frozen target, and an issue comment with the revision, backend versions, entry point, exact steps and observed result. | `feature-map:not-reproduced` | Amber `#D4A72C` |+| Out of scope | A report row explaining why the issue has no current mapped Canvas feature. | `feature-map:out-of-scope` | Gray `#6E7781` |++These are accounting outcomes. A mapped issue is not necessarily reproduced;+its note distinguishes a reproduced failure from a reported failure blocked by+a prerequisite. Not reproduced does not mean fixed, and out of scope does not+dismiss the bug. Record verification separately: a `Known issue` with an+unavailable prerequisite is **BLOCKED**, with that prerequisite named; it is not+a `Known failure`. Reserve `Known failure` for an observed live reproduction+and preserve its date. An available scenario awaiting a run is `not-run`, not+blocked merely because it has not been attempted. Text carries this meaning+even when label colors cannot be seen. Each processed issue has exactly one outcome label in repositories where+these labels have been authorized. An unprocessed issue has none; keep it in+the pending section of the ledger instead of guessing a completed outcome.++## Evidence and labels++The map note or report is the record; GitHub labels are a searchable mirror.+Create labels or publish comments only when authorized by the requester. Before+adding a label, validate its supporting note or run, then remove only any of the+other two outcome labels on that same issue. Preserve all unrelated labels.+Cross-repository notes spell out the repository in their issue links.++A successful build, source fix or unit test cannot establish not reproduced.+Run the original user-visible scenario through `control-openhands`, and capture+its read-only result as the recipe requires. If the reproduction can be driven,+run it; do not turn a time limit into a missing-prerequisite claim. When no+recipe can drive a reported scenario, record the exact missing account, model,+OS, old persisted data or reproduction detail in the Gotchas note. Preserve+reproduction dates for unresolved notes already in the map. The map describes+current application behavior: once a fix applies at the current Canvas and+pinned backend versions, remove the resolved failure note and obsolete+reproduction recipe. Keep earlier observations in the dated backfill ledger,+not in historical sections of the feature map.++For a not-reproduced comment, introduce the agent and requester as instructed,+then include the frozen revision, actual backend versions, run ID, user entry+point, commands, expected behavior and observation. Leave closure to maintainers.+Do not publish keys, live session URLs, raw traces or unreviewed screenshots.++## Pass ledger++Keep one row per `(repository, issue number)` with outcome, owning family and+feature IDs, note or comment link, verification status, exact missing+prerequisite when blocked, and run/evidence details+when exercised. Report counts per repository and distinguish existing notes,+new reproductions and blocked known issues within the mapped outcome. Include+all pending rows, even if another shard owns them. The pass is complete only+when every frozen-inventory issue has exactly one supported outcome, every+not-reproduced comment has been published, and `control-openhands map check`+reports `ok`. Follow [report.md](report.md) for run identity, coverage and+teardown, and [daily.md](daily.md#several-agents) for whole-family sharding.++Daily and weekly passes query open bugs without a supported map or ledger+outcome, and re-triage reopened or newly bug-labeled issues even if an older+snapshot recorded an outcome. A historical ledger does not by itself cover a+changed scenario. Do not use issue creation time or `ready-for-dev` as a+completeness filter.++Re-check a not-reproduced result when its scenario, relevant source or backend+changes; a result applies to its recorded revision and configuration. Re-check+out-of-scope classification if the feature becomes supported. After a mapped+issue closes, confirm the fix applies at the current Canvas and pinned backend+versions and re-drive its scenario where possible. Remove resolved bug notes+and obsolete recipes from the map; retain the current expected behavior.+Record verification limits and earlier observations in the dated ledger. Keep+label reconciliation in that same pass.++The completed [2026-10-12 backfill](backfills/2026-10-12-open-bugs.md) records+its frozen inventory, per-issue outcomes, published comments and retained run+evidence. The [2026-10-16 main refresh](backfills/2026-10-16-main-refresh.md)+records the later inventory and reconciles closed reports with current source,+pinned backend versions and fresh observations. Read the latest reconciliation+before treating a historical result as current. Fixture tests use the installed SDK Python, which supplies `httpx`:+`python -m unittest discover -s .agents/skills/verify-openhands/scripts/fixtures -p 'test_*.py'`.++## Refresh after main changes++Keep the original snapshot and its evidence dated. Freeze the new main target+and issue inventory separately, inherit main's accepted maintenance baseline,+and add a reconciliation ledger. Re-read every affected family, including+notes imported by main and closed issues absent from the original snapshot.+A merged SDK fix may be newer than Canvas's released pin: a closed issue is+not enough to retire a supported-version failure. Conversely, when main has+fixed or replaced a flow, describe the current behavior and remove obsolete+failure notes and reproduction steps. Keep superseded observations only in+the dated backfill ledger, outside the current feature map.++For each closed report distinguish current live observation, accepted source+change awaiting the original scenario, and a fix unavailable at the tested+pin. A source review, closure or passing unit test is not a verified runtime+fix. Retain the concrete account, model, platform or fixture prerequisite for+scenarios that cannot be driven. Refresh changed not-reproduced scenarios and+append the new revision, pins and steps to their existing issue comments; keep+the earlier result intact. Reconcile only the three outcome labels after the+current supporting record exists, leaving unrelated labels unchanged.diff --git a/.agents/skills/verify-openhands/references/daily.md b/.agents/skills/verify-openhands/references/daily.mdindex f398ecf269..d60ccd84f4 100644--- a/.agents/skills/verify-openhands/references/daily.md+++ b/.agents/skills/verify-openhands/references/daily.md@@ -1,7 +1,7 @@# The daily passThe weekly pass in [maintenance.md](maintenance.md) proves every feature live.-A daily pass cannot: 27 families and 700-odd sub-features take longer than one+A daily pass cannot: 28 families and 700-odd sub-features take longer than oneagent's day, and most of the map did not change since yesterday. The dailypass answers a narrower question with the same rigor: **did anything thatmerged since the last pass change what a user sees, and does the rest still@@ -72,12 +72,16 @@ cheaper than the next and catches a different kind of rot.results, as [maintenance.md](maintenance.md) step 2 describes. The deltapass has a second half: every bullet whose Gotchas or Known failure notelinks an issue that closed since `BASE` (check each linked issue's state)- is re-driven too, its row recorded and its note updated (a note with a- `known-bug` block is re-driven with `control-openhands bug run <ID>`; an- `xpass` goes to map review per [maintenance.md](maintenance.md) step 5). A fix that landed- in the Agent Server or the automation service shows up only this way. So is- every open `bug` issue opened since `BASE` that names a mapped feature: it- is reproduced and gets its note, so bugs filed outside a pass reach the map.+ is re-driven too, its row recorded and its note updated. Re-drive a+ `known-bug` block with `control-openhands bug run <ID>`; an `xpass` goes to+ map review per [maintenance.md](maintenance.md) step 5. A fix in the Agent+ Server or automation service shows up only this way. Query open `bug`+ issues without a supported map or pass-ledger outcome, regardless of creation+ date, and re-triage reopened or newly bug-labeled reports. Drive actionable+ reports and add their note or recorded not-reproduced result. Include older+ actionable reports: `ready-for-dev` sets priority, not inventory membership.+ Reconcile outcome labels against their note or recorded run per+ [bug-backfill.md](bug-backfill.md); labels alone are not reproduction evidence.3. **Smoke (every family, cheaply).** For each family not already driven intier 2: open its first entry point, run the family's `errors --app-only`sweep, and drive the one bullet that proves the page's main state (a list@@ -90,7 +94,7 @@ cheaper than the next and catches a different kind of rot.the date, not from yesterday's notes: with `d` the UTC day of the year,today's families are the ones at positions `(d mod 7) * 4` to`(d mod 7) * 4 + 3` of the index table, in ID order (position 24 onwards- holds F25 to F27). Model-backed bullets run only inside the LLM budget;+ holds F25 to F28). Model-backed bullets run only inside the LLM budget;without a key they are `blocked` with the missing prerequisite.## Reading the resultdiff --git a/.agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md b/.agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.mdindex 01dde165f9..abb0bd3c33 100644--- a/.agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md+++ b/.agents/skills/verify-openhands/references/feature-map/F01-first-run-and-sign-in.md@@ -104,9 +104,11 @@ Not reachable locally:- The consent modal (z-70) sits over the onboarding modal, and clicks at the screen center land on agent tiles once it closes. A forced click on `first-run-onboarding-screen` selected Codex. Check backdrop non-dismissal with `browser mouse-click 40 40`, a corner outside the modal.- The say-hello default message ("Create a basic webpage…") makes the agent work for a while. Replace it with a tiny prompt to keep model cost down.- On a local backend, Next on the LLM step also points the `default` agent profile at the profile it just created. Activating another LLM profile does not move that pointer, and while it stays, deleting the onboarding profile answers `409` `LLM profile is referenced by 1 agent profile(s): default`. `control-openhands llm preset deepseek` moves it to `deepseek-flash` and reports `repointed`; `llm set` does not. In the UI, edit `default` in Settings → Agents and pick the profile in `testid=agent-profile-llm-selector`.-- On a local backend the LLM step edits an LLM profile and never reads or writes the raw LLM settings (#18218, after #17889 fixed #17884). It takes the active profile, else the one the `default` agent profile points at (the Agent Server's backfill of an older install's raw settings), else a fresh one with the default model. It reads the profile only once the backend's health check reports it connected, so a step mounted while the Agent Server was down fills in after it comes back (verified 2026-10-12 with `control-openhands service stop agent-server`, then `restart`). If the profiles cannot be read it shows `Failed to load existing profile. Please try again.` (`testid=onboarding-llm-load-error`) with Retry (`testid=onboarding-llm-load-retry`) and Next disabled. Next without changes only sends `POST /api/profiles/<name>/activate`.+- On a local backend the LLM step edits an LLM profile and never reads or writes the raw LLM settings (#18218). It takes the active profile, else the one the `default` agent profile points at (the Agent Server's backfill of an older install's raw settings), else a fresh one with the default model. It reads the profile only once the backend's health check reports it connected, so a step mounted while the Agent Server was down fills in after it comes back (verified 2026-10-12 with `control-openhands service stop agent-server`, then `restart`). If the profiles cannot be read it shows `Failed to load existing profile. Please try again.` (`testid=onboarding-llm-load-error`) with Retry (`testid=onboarding-llm-load-retry`) and Next disabled. Next without changes only sends `POST /api/profiles/<name>/activate`.- The tab title prefixes a status emoji to the stored title. A generated title that already starts with an emoji therefore shows twice (`✅ ✅ Reply hello without tools | OpenHands`). Right after launch the title is `Conversation <id prefix>` until a reload picks up the generated one.- Right after Next, the ACP step reads "Checking for an existing Claude Code login…" with Next disabled. Wait for `testid=onboarding-acp-auth-detected` before reading the banner. The "already signed in" banner depends on the host's CLI login. On a clean CI machine the fields become required, and Next is blocked until they are filled.- Choosing an ACP agent saves `agent_kind: acp` immediately. Re-choose OpenHands before leaving, or later conversations in the run use the ACP agent.- In preview mode, Skip and Close are inert. The LLM slide shows `gpt-5.6-sol` even when another model is saved, so do not read it as data loss.- `browser toasts --history` lists every status or alert text (toasts and alert banners) seen since the page loaded, including success toasts and progress lines such as `Checking your backend connection…`. Read it on a page that has just loaded (or reloaded) to check that no error toast appeared. A page with a standing banner never reads `[]`: the app shell of a backend with no LLM lists the `Your LLM isn't set up yet` alert.+- Known issue [#16601](https://github.com/OpenHands/OpenHands/issues/16601) (`F01.telemetry-consent`): Analytics consent reappears sporadically. Live prerequisite: The multi-day trigger and backend consent response when the modal returns; opted-in localStorage alone does not reproduce it.+- Known issue [#16526](https://github.com/OpenHands/OpenHands/issues/16526) (`F01.onboarding-choose-agent`): 403 error when selecting an agent in the onboarding agent selection modal. Live prerequisite: An authenticated hosted Cloud account with the reported organization-membership mismatch during onboarding.diff --git a/.agents/skills/verify-openhands/references/feature-map/F02-app-shell.md b/.agents/skills/verify-openhands/references/feature-map/F02-app-shell.mdindex 6b1aa024b9..e329672b36 100644--- a/.agents/skills/verify-openhands/references/feature-map/F02-app-shell.md+++ b/.agents/skills/verify-openhands/references/feature-map/F02-app-shell.md@@ -93,6 +93,8 @@ Preconditions:- **Check for errors.** Run `control-openhands browser errors --app-only` after each group; it reports `pageErrors` `0` and `appErrors` `0`. External `registry.npmjs.org` and `cdn.simpleicons.org` failures are sandbox noise.- **Restore.** Leave the sidebar expanded, unpinned, with the checklist shown and expanded, at `browser viewport desktop`.+- **Existing conversations while collapsed (`F02.sidebar-collapse`).** With at least one sidebar conversation, run `control-openhands browser click 'testid=sidebar-collapse-toggle'`, `control-openhands browser snapshot` and `control-openhands browser count 'testid=conversation-panel'`. Known failure (reproduced 2026-10-12, [#15662](https://github.com/OpenHands/OpenHands/issues/15662)): the count is `0`, and the rail has no chooser for existing conversations; expansion is required. Restore with `control-openhands browser hover 'role=link[name="OpenHands Logo"]'` and `control-openhands browser click 'testid=sidebar-collapse-toggle'`.+## Gotchas- The rail animates its width for 200 ms. A `bbox` taken right after the collapse click reads an intermediate width (about 70); measure after `browser reload`.diff --git a/.agents/skills/verify-openhands/references/feature-map/F03-home.md b/.agents/skills/verify-openhands/references/feature-map/F03-home.mdindex a4a91c70df..d0472619a6 100644--- a/.agents/skills/verify-openhands/references/feature-map/F03-home.md+++ b/.agents/skills/verify-openhands/references/feature-map/F03-home.md@@ -121,7 +121,7 @@ Automations on home (fixtures imported and turned on):- **Missing integration (`F03.recommended-missing-integration`).** `control-openhands browser click 'testid=recommended-automation-rail-card-research-brief-writer'` opens `dialog "Tavily"` (Command `npx -y tavily-mcp`, Tavily API key); Cancel closes it and no conversation is created.- **Already added (`F03.recommended-hides-added`).** `control-openhands browser count 'testid=recommended-automation-rail-card-news-digest'` is `1`. Import `qa-f03-news.automation.json` (same fixture recipe as Preconditions, with `"name":"Daily news digest"`; it stays disabled), `browser goto /`, `browser wait 'testid=recommended-automation-rail-card-custom-automation'`: the news-digest count is `0`. Delete it on `/automations` (`control-openhands browser goto /automations`, then the Cleanup recipe below with `has-text=Daily news digest`); back on `/` (`control-openhands browser goto /`) the count is `1` again.- **Activity list (`F03.home-automations-list`, `F03.home-automation-links`).** `control-openhands browser wait '[data-testid^="running-automation-row-"]'`, `control-openhands browser text 'testid=running-automations-list'`: both fixtures with `0 0 1 1 *` and `No activity yet` (`browser screenshot --feature F03.home-automations-list --name list`); no row has a sparkline link yet (the run-bar link is checked after the first run, in the next bullet). `control-openhands browser click '[data-testid^="running-automation-row-"] >> has-text=QA_F03 Beta >> role=link' --expect-url '/automations/[^/?]+'` opens the detail page; after `control-openhands browser goto /`, `control-openhands browser click 'testid=home-automations-manage' --expect-url '/automations$'` opens the dashboard.-- **Row menu, Run now and Cancel (`F03.home-row-menu`, `F03.cancel-in-flight-run`, `F03.home-automation-links`).** Run `control-openhands browser goto /` (the Manage link left the dashboard), then `control-openhands browser click '[data-testid^="running-automation-row-"] >> has-text=QA_F03 Pong >> [data-testid^="running-automation-menu-"]'` and `control-openhands browser snapshot '[data-testid^="running-automation-menu-panel-"]'`: Run now, View, Edit, Turn off, separator, Pin to dashboard. `control-openhands browser click '[data-testid^="running-automation-run-"]' --observe 'role=status'` observes `Automation dispatched`. Reopen the menu: `Run now [disabled]` and **Cancel run**. `control-openhands browser click '[data-testid^="running-automation-cancel-"]'` and `control-openhands browser toasts` reads `Automation run cancelled`; `control-openhands api GET /api/automation/v1/<automation-id>/runs` (id from `browser testids --filter running-automation-row`) shows `CANCELLED`, and after `browser reload` the row reads `Just now · Cancelled`. Pong now has a run: `control-openhands browser click '[data-testid^="running-automation-activity-"] >> role=link' --expect-url '/automations/[^/?]+\?run='` opens it; return with `control-openhands browser goto /`.+- **Row menu, Run now and Cancel (`F03.home-row-menu`, `F03.cancel-in-flight-run`, `F03.home-automation-links`).** Run `control-openhands browser goto /` (the Manage link left the dashboard), then `control-openhands browser click '[data-testid^="running-automation-row-"] >> has-text=QA_F03 Pong >> [data-testid^="running-automation-menu-"]'` and `control-openhands browser snapshot '[data-testid^="running-automation-menu-panel-"]'`: Run now, View, Edit, Turn off, separator, Pin to dashboard. `control-openhands browser click '[data-testid^="running-automation-run-"]' --observe 'role=status'` observes `Automation dispatched`. Reopen the menu: `Run now [disabled]` and **Cancel run**. `control-openhands browser click '[data-testid^="running-automation-cancel-"]'` and `control-openhands browser toasts` reads `Automation run cancelled`; `control-openhands api GET /api/automation/v1/<automation-id>/runs` (id from `browser testids --filter running-automation-row`) shows `CANCELLED`, and after `browser reload` the row reads `Just now · Cancelled`. Pong now has a run: `control-openhands browser click '[data-testid^="running-automation-activity-"] >> role=link' --expect-url '/automations/[^/?]+\?run='` opens it; return with `control-openhands browser goto /`. Known failure (reproduced 2026-10-12, [OpenHands/automation#514](https://github.com/OpenHands/automation/issues/514)): a bounded script run cancelled from this Home menu shows `Automation run cancelled` and `Cancelled` in the row and Activity Log, but its launcher-owned `python3 worker.py` process remains running after cancellation. This was reproduced in `backfill-18309-integrations-worker` on `fa1f724ab5737a17453ad6081129eb91c5a423c6`, SDK 1.54.0 and automation 1.19.3: package a run-owned `worker.py` that prints one tick and sleeps 2 seconds for 90 iterations; upload its tarball with `control-openhands api POST '/api/automation/v1/uploads?name=qa-backfill-sleep' --write --data-file <bounded-sleep-tarball> --content-type application/gzip`, then create an active scripted automation through `api POST /api/automation/v1 --write --data '@<fixture-json>'` with `entrypoint: python3 worker.py`, the returned `tarball_path`, a UTC `0 0 1 1 *` cron trigger and timeout 240. Open `/automations/<sleep-id>`, click `role=button[name="Run now"]`, wait for the actual worker to start, return Home and click `testid=running-automation-menu-<sleep-id>` then `testid=running-automation-cancel-<sleep-id>`. Read-only process checks restricted to this launcher's descendants still found the worker after the cancellation UI settled. Evidence: `F03.cancel-in-flight-run/local-worker-cancel-514.png`, `F03.cancel-in-flight-run/owned-worker-post-cancel.json` and `F23.activity-log/cancelled-514.aria.txt`; the worker is bounded to exit without external calls. Let it exit, then delete the temporary automation with `control-openhands api DELETE /api/automation/v1/<sleep-id> --write`.- **Tooltip (`F03.home-run-tooltip`).** `control-openhands browser tooltip '[data-testid^="running-automation-row-"] >> has-text=QA_F03 Pong'` returns `QA_F03 Pong`, `Trigger 0 0 1 1 *`, `Status Cancelled`, `Last run Just now`, `Task Cancelled by user`.- **Run outcomes (`F03.home-run-outcome-health`).** Arrange as `F23.run-task-outcome` does (API `preset/prompt`, yearly cron `0 0 1 1 *`, enabled) `QA_F03 Blocked`, which finishes with status `blocked`, and `QA_F03 Review`, which finishes with status `unknown`, each prompt ending `Do not run any tools.` On `/`, for each row open `'[data-testid^="running-automation-row-"] >> has-text=QA_F03 Blocked >> [data-testid^="running-automation-menu-"]'` and click `[data-testid^="running-automation-run-"]` (`Automation dispatched`); poll `control-openhands api GET /api/automation/v1/<id>/runs` until the run is `COMPLETED` (about 35 s), then `control-openhands browser reload`. `control-openhands browser text 'testid=running-automations-list'` shows `… · Blocked` and `… · Needs review`; `control-openhands browser eval "[...document.querySelectorAll('[data-testid^=running-automation-row-] a[aria-label]')].filter(a=>a.querySelector('.rounded-full')).map(a=>a.getAttribute('aria-label')+' | '+a.querySelector('.rounded-full').className.match(/bg-[a-z-]+/g)).join('; ')"` lists `QA_F03 Blocked Blocked | bg-warning` and `QA_F03 Review Needs review | bg-warning` (in list order, which can vary). `control-openhands browser tooltip` on each row reads `Status Blocked` / `Task <its summary>` and `Status Needs review` / `Task <its summary>` (before each tooltip read, `browser hover 'testid=chat-input'` and `browser wait 'role=tooltip' --state detached --timeout 3000`, or the previous row's tooltip is returned). Pin both from their row menus (`[data-testid^="running-automation-pin-"]`): each card's strip shows the Blocked (!) (`run-status-icon-warning`) or Needs review (?) (`run-status-icon-needs-review`) icon, then the run's summary and its conversation title. Expected: the summary is visible (`browser eval "[...document.querySelectorAll('[data-testid^=pinned-automation-card-] [data-testid=run-status-badge-icon]')].map(i=>Math.round(i.nextElementSibling.getBoundingClientRect().width))"` is above `0`). Known failure (reproduced 2026-10-09, #18234): when the run has a conversation, the summary is 0 px wide at desktop and phone, so only the conversation title shows (the summary is still in `browser text`). Repeat at `browser viewport phone`. The status-less success case has no live producer (automation 1.19.0 with SDK 1.53.0 always sends a status), as `F23.run-task-outcome` records: `blocked`. Delete both automations afterwards.- **View and Edit (`F03.home-row-menu`).** From the Beta row menu, `control-openhands browser click '[data-testid^="running-automation-view-"]' --expect-url '/automations/[^/?]+$'` opens the detail page. Back on `/` (`control-openhands browser goto /`), reopen the menu and click `[data-testid^="running-automation-edit-"]`, then `control-openhands browser wait-url 'automations/setup\?automationId='`: the setup page opens prefilled (no modal; `browser count 'text=Edit automation'` is `0`), `browser value 'testid=automation-setup-name'` is `QA_F03 Beta` and the prompt is the stored one (`browser screenshot --feature F03.home-row-menu --name edit-opens-setup`). `control-openhands browser fill 'testid=automation-setup-prompt' 'Reply with the single word: beta2. Do not run any tools.'` and `control-openhands browser click 'testid=automation-setup-save' --observe 'role=status'` (`Automation updated`); `api GET /api/automation/v1/<automation-id>` shows the new prompt. The page's own controls (Test, Close, the timeout limit) are F22's.@@ -137,6 +137,13 @@ Phone and cleanup:- **Phone (`F03.phone`).** `control-openhands browser viewport phone`, `control-openhands browser bbox 'testid=pinned-automations-dashboard'` and `... 'testid=running-automations-list'` (358 px wide, `pageHorizontalOverflow` `false`), `browser screenshot --feature F03.phone --name automations` (the banner was checked at phone width in the No-LLM banner bullet). Open Workspace: `control-openhands browser click 'testid=open-workspace-button'`, then `browser bbox 'testid=open-workspace-dialog-body'` is inside the viewport. Folder browser: arrange a long name first with `control-openhands fixture folder --name qa-f03-folder-with-a-name-too-long-for-a-phone` (it lands in `<run>/workspace`), then `control-openhands browser click 'testid=workspace-dropdown'` and `control-openhands browser click 'testid=add-workspaces-button'`; `control-openhands browser eval "(()=>{const b=document.querySelector('[data-testid=folder-browser-entry-workspace] span span.truncate');return b.getBoundingClientRect().width})()"` is greater than `0` (about 67). Run `control-openhands browser click 'testid=folder-browser-up'` twice (now `<run>`) and `control-openhands browser click 'testid=folder-browser-entry-workspace'`, then `control-openhands browser eval "(()=>{const b=document.querySelector('[data-testid=folder-browser-entry-qa-f03-folder-with-a-name-too-long-for-a-phone] span span.truncate');return [b.scrollWidth>b.clientWidth,getComputedStyle(b).textOverflow]})()"`: `[true,"ellipsis"]` (about 109 of 293 px shown). `control-openhands browser screenshot --feature F03.phone --name folder-names` shows each name beside `Folder` under separate `Name` and `Kind` headers, the long one cut to `qa-f03-folder-w…`. Close with `control-openhands browser click 'testid=folder-browser-cancel'` and `control-openhands browser click 'testid=close-open-workspace-dialog'`, then return with `control-openhands browser viewport desktop`.- **Cleanup (`F03.pinned-automations-grid`).** Delete both fixtures on `/automations` (`control-openhands browser goto /automations`; F21: kebab `role=list >> has-text=Export >> role=button[name="Delete"]`, then `role=heading[name="Delete automation"] >> xpath=.. >> role=button[name="Delete"]`). On `/` (`control-openhands browser goto /`), `browser count 'testid=pinned-automations-dashboard'` is `0` and `browser text 'testid=running-automations-empty-hint'` is `No automations yet. Add one to automate repetitive work.`+- **Implicit-launch readiness (`F03.composer-blocked-without-llm`).** Arrange named OpenHands agent `QA_PINNED` with `tools:[]`, concurrency7, and two LLM profiles on refused loopback `http://127.0.0.1:1/v1`: `QA_READY` has a disposable key and `QA_EMPTY` has no key. Use real `/api/profiles/<name>` and `/api/agent-profiles/<name>` writes, then activate the named agent by its ID. Pin it to QA_EMPTY and activate QA_READY; open Home and inspect the warning/pill. Invert the pinned/active profiles, reopen Home, fill `testid=chat-input` with a bounded no-tools prompt and read `browser enabled testid=submit-button`. Expected: readiness agrees with the effective LLM shown by the launch pill. Known failure (reproduced 2026-10-12, [#16951](https://github.com/OpenHands/OpenHands/issues/16951)): the readiness gate follows the named agent's pinned LLM, so a keyless pin can disable a key-present pill, and the inverse can permit a keyless pill. On current backends with per-launch LLM override, a launch uses the pill's LLM while preserving the named agent's identity, concurrency and tools (`F13.llm-pill-precedence`); that current behavior is separate from the readiness mismatch. Read only credential-presence booleans, without claiming provider authentication. Restore the default agent/active LLM and delete disposable profiles.++- **Target across Local backends (`F03.selection-after-reload`).** In two run-owned Local backends, save distinct QA_A_ONLY and QA_B_ONLY workspaces through real `/api/workspaces` writes. Register the second backend through Add Backend with its key read from its private key file. On QA_Secondary Home, open the workspace dialog, choose QA_B_ONLY and Confirm; switch to Local through `backend-selector` without reload or `browser goto`. Read the preview, submit a bounded no-tools prompt through the real composer, and read the new Local conversation workspace. Expected: an incompatible target is cleared before submission. Known failure (reproduced 2026-10-12, [#16950](https://github.com/OpenHands/OpenHands/issues/16950)): QA_B_ONLY remains visible and its secondary-run path is submitted to Local. This is the Local variant; Cloud repository/organization recents still need an approved Cloud fixture.+- **Saved workspaces across Local backends (`F03.workspace-dropdown-search`).** Warm the Local picker with QA_A_ONLY, close its dialog, switch through `backend-selector >> role=option[name="Connected QA_Secondary"]`, then reopen the picker within60seconds, without reload or hard navigation. Expected: only QA_B_ONLY from the active server registry is offered. Known failure (reproduced 2026-10-12, [#16844](https://github.com/OpenHands/OpenHands/issues/16844)): the recorded2.7-second switch keeps QA_A_ONLY in the secondary picker while independent API readbacks show distinct registries. The same-absolute-parent directory variant requires isolated filesystems; two native runs share a host filesystem, so it and Windows remain unverified. These checks used frozen target `fa1f724ab5737a17453ad6081129eb91c5a423c6`, Agent Server/SDK1.54.0, automation1.19.3, run `backfill-18309-coordinator-checks` plus its owned `backfill-18309-local-secondary`. Retain `evidence/F03.selection-after-reload/16950-submission-readback.json`, `evidence/F03.workspace-dropdown-search/16844-local-switch.mp4` and registry readback; remove the registered backend/workspaces and stop both runs afterward.++- **Invalid project sub-agent (`F03.launch-in-workspace`).** In an owned workspace create `.agents/agents/qa-bad-skill.md` with YAML frontmatter `name: qa-bad-skill`, `description: Bad skill`, `skills: qa-f27-missing-skill`, then body `Body`, without installing that skill. Use an explicit dummy profile pointed at a refused loopback endpoint with autotitle off for this pre-inference check. On Home use `control-openhands workspace open <owned-folder>`, select `testid=workspace-mode-selector-option-local_repo`, fill `testid=chat-input` and click `testid=home-chat-launcher >> testid=submit-button` with `--expect-url "/conversations/[^/]+$"`. The invalid sub-agent is skipped with a warning; creation returns201 and opens its conversation instead of500. A later refused-loopback model error is separate from creation. Read `browser network --filter "api/conversations($|\?)"`, capture `browser screenshot --feature F03.create-error-toast --name invalid-subagent`, then remove only the owned bad-agent file.+## Gotchas- Once a workspace is confirmed, `open-workspace-button` is not rendered (the preview bar replaces it): a click on it waits the full 30 s. Reopen through `home-git-control-bar-preview`, or reload.@@ -157,3 +164,11 @@ Phone and cleanup:- The Code / Automate choice is component state, not stored: every page load and every return to home starts in Code. A toggle check after a navigation must click Automate again.- When the Agent Server is stopped, background refetches can add a generic `An error occurred` toast (#18051). For `F03.create-error-toast`, retain the complete video and toast history and report any extra toast with its timing; do not waive it from the count.- The workspace combobox uses one message for an empty list and for a search with no match: both read `No workspaces yet` (`workspace-dropdown-empty`), with **+ Add Workspace** still below it (#18037).+- Known issue [#17108](https://github.com/OpenHands/OpenHands/issues/17108) (`F03.open-repository-dialog`): The repository picker lists a search result for input the user has already cleared. Live prerequisite: An authenticated Cloud Git provider and a real repository search response that remains pending after clearing the input.+- Known issue [#16801](https://github.com/OpenHands/OpenHands/issues/16801) (`F03.launch-without-workspace`): Agent starts by looking in workspace, even when started with "no workspace". Live prerequisite: A working Claude Opus agent and the original task/prompt that makes it explore the default workspace; the prompt is absent.+- Known issue [#16210](https://github.com/OpenHands/OpenHands/issues/16210) (`F03.plugin-picker`): [Cloud Backends] Plugins not available when starting a new conversation. Live prerequisite: An authenticated Cloud organization with configured account plugins before opening New Chat > Plugins.+- Known issue [#15701](https://github.com/OpenHands/OpenHands/issues/15701) (`F03.launch-in-workspace`): Creating Conversation toast stuck visible. Live prerequisite: The original stuck creation request/outcome sequence; the report is intermittent and gives no deterministic trigger.+- Known issue [#18480](https://github.com/OpenHands/OpenHands/issues/18480) (`F03.open-repository-dialog`): an Azure DevOps Enterprise user belonging to several organizations can see only one organization's repositories. Live prerequisite: a real Cloud/Enterprise sign-in with two organizations and distinct repositories; the report's mocked two-organization capture is not a real-backend reproduction.+- Known issue [#18418](https://github.com/OpenHands/OpenHands/issues/18418) (`F03.create-error-toast`, `F03.launch-with-plugin`): hosted runtime 1.50.1 can fail during skills/plugins setup because its packaged plugin-schema resource is missing. Live prerequisite: the affected Cloud runtime/PyInstaller image and original plugin setup plus its account. A Local 1.55.0 launch does not verify that deployment.+- Known issue [OpenHands/software-agent-sdk#5837](https://github.com/OpenHands/software-agent-sdk/issues/5837) (`F03.folder-browser`): case-only directory names at a page boundary can repeat the server cursor; Canvas detects the repeated cursor and the directory listing fails instead of showing all choices. Live prerequisite: a case-sensitive filesystem with enough directories to cross the real picker page boundary, including two distinct case-only names. A case-insensitive Mac volume cannot hold that fixture.+- Known issue [OpenHands/software-agent-sdk#5849](https://github.com/OpenHands/software-agent-sdk/issues/5849) (`F03.folder-browser`, `F03.manage-workspaces`): re-adding a saved workspace or removing an already-removed stale row can return500 because the registry writes even for a no-op. Live prerequisite: an owned read-only workspace registry plus the already-saved or stale picker row; changes to registry mtime/inode alone are not a Canvas failure.diff --git a/.agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md b/.agents/skills/verify-openhands/references/feature-map/F04-conversation-list.mdindex a27d790ade..e50608ead1 100644--- a/.agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md+++ b/.agents/skills/verify-openhands/references/feature-map/F04-conversation-list.md@@ -79,7 +79,7 @@ Preconditions:- **Empty list (`F04.empty-state`).** On a run with no conversations, run `control-openhands browser text 'testid=conversation-panel-empty-state'`. The text is `No conversations found`.- **View menu (`F04.view-menu`).** Run `control-openhands browser click 'testid=conversation-layouts-toggle'` and `control-openhands browser snapshot 'testid=conversation-layouts-menu'`: radios `By workspace`, `Recent` (`[checked]`), `Active`, `Compact`, items `Filter by tags`, `More options` and `Delete all` (`control-openhands browser attr 'testid=layout-preset-recent-activity' aria-checked` is `true`). With no conversations `control-openhands browser enabled 'testid=delete-all-conversations'` is `false`; after `control-openhands browser click 'testid=tag-filters-section'`, `control-openhands browser text 'testid=tag-filters-empty'` is `No tags yet`. `control-openhands browser press Escape` closes the menu (`browser count 'testid=conversation-layouts-menu'` is `0`).-- **Start in a workspace (`F04.new-thread-picker`).** Run `control-openhands browser goto /`, `control-openhands browser click 'testid=conversation-panel-new-thread-picker'` and `control-openhands browser snapshot 'testid=conversation-panel-new-thread-popover'`: buttons `No workspace`, `qa-f04-folder`, `qa-f04-repo`, `+ Add Workspace`, `Manage Workspaces`. Then `control-openhands browser click 'testid=conversation-panel-new-thread-popover >> testid=launch-workspace >> has-text=qa-f04-repo' --expect-url '/conversations/'` returns `/conversations/<id>`, and `control-openhands conversation list` shows that id with `workspace` ending in `/workspace/qa-f04-repo`. From `/` again, `testid=launch-no-workspace` opens a new conversation whose row sits in the No workspace folder. `control-openhands browser click 'testid=manage-workspaces-button'` opens `testid=manage-workspaces-modal` (F03); Escape closes it.+- **Start in a workspace (`F04.new-thread-picker`).** Run `control-openhands browser goto /`, `control-openhands browser click 'testid=conversation-panel-new-thread-picker'` and `control-openhands browser snapshot 'testid=conversation-panel-new-thread-popover'`: buttons `No workspace`, `qa-f04-folder`, `qa-f04-repo`, `+ Add Workspace`, `Manage Workspaces`. Then `control-openhands browser click 'testid=conversation-panel-new-thread-popover >> testid=launch-workspace >> has-text=qa-f04-repo' --expect-url '/conversations/'` returns `/conversations/<id>`, and `control-openhands conversation list` shows that id with `workspace` ending in `/workspace/qa-f04-repo`. From `/` again, `testid=launch-no-workspace` opens a new conversation whose row sits in the No workspace folder. `control-openhands browser click 'testid=manage-workspaces-button'` opens `testid=manage-workspaces-modal` (F03); Escape closes it. Immediately after creating that empty workspace conversation, `control-openhands conversation status <id>` reads `idle`; inspecting the `conversation-status-active` class shows a green dot with no animation class. Known failure (reproduced 2026-10-12, [#16752](https://github.com/OpenHands/OpenHands/issues/16752)): this empty idle row retains active green styling instead of the expected neutral state; the older report's flashing animation was not observed.- **Row opens its conversation (`F04.card`).** From `/` (`control-openhands browser goto /`; the previous bullet's launch left a conversation page, where `--expect-url '/conversations/'` would pass at once) run `control-openhands browser click 'testid=conversation-card >> has-text=pong' --expect-url '/conversations/'`, then `control-openhands browser attr 'testid=conversation-card >> has-text=pong' data-active` (`true`; other rows `false`). `control-openhands browser eval "document.querySelector('[data-testid=conversation-card][data-active=true]').closest('a').getAttribute('href')"` is `/conversations/<id>?backend=default-local`. The finished conversation shows `testid=conversation-status-check`, idle ones `conversation-status-active` (green), paused ones `conversation-status-paused` (grey). The reopened chat has its transcript back: `control-openhands browser wait 'testid=agent-message'`, then `control-openhands browser text 'testid=agent-message >> nth=-1'` is `pong` and `control-openhands browser text 'testid=user-message >> nth=-1'` is the prompt, with `control-openhands browser count 'testid=error-message-banner'` `0` (`browser screenshot --feature F04.card --name reopened`).- **Live status (`F04.card-live-status`).** Run `control-openhands conversation start --prompt "Reply with only the word ping. Do not run any tools."` (no `--wait`; one tiny model call), then `control-openhands browser wait 'testid=conversation-panel >> testid=conversation-status-working' --timeout 20000`: the active row (`Conversation <id5>`) shows the working indicator. After `control-openhands conversation wait <id> --timeout 180` reports `finished`, `control-openhands browser wait 'testid=conversation-panel >> [data-testid=conversation-card][data-active=true] >> testid=conversation-status-check' --timeout 60000` succeeds and `control-openhands browser text 'testid=conversation-panel >> [data-testid=conversation-card][data-active=true] >> testid=conversation-card-title'` is the generated title (it varies: `✅ Reply with only ping`, `✅ Reply only ping, no tools`). The row lags the conversation by up to about 40 s (list refresh).- **Group by workspace (`F04.grouped-folders`).** Run `control-openhands browser click 'testid=conversation-layouts-toggle'`, `control-openhands browser click 'testid=layout-preset-by-workspace'`, then `control-openhands browser snapshot 'testid=conversation-panel'`. Regions `No workspace`, `qa-f04-repo` and `qa-f04-folder` each list their own conversations; a workspace without conversations still gets a folder. Second view in another browser: `control-openhands browser reset`, `control-openhands onboard --skip`, choose By workspace again; the workspace conversations must still be in their folders (currently they all fall into No workspace, a known failure).@@ -114,9 +114,12 @@ Preconditions:- **Show archived and Unarchive (`F04.show-archived-unarchive`).** In More options click `testid=toggle-show-archived` (`aria-checked` `true`) and `testid=advanced-options-close`; `control-openhands browser count 'testid=conversation-card >> has-text=Conversation <id5> >> testid=conversation-card-archived-chip'` is `1` and the ⋮ menu offers `Unarchive Conversation`. `control-openhands browser click 'testid=context-menu >> testid=unarchive-button'` opens no dialog (`browser count 'role=dialog'` `0`) and the chip count drops to `0`. Turn Show archived off again.- **Delete (`F04.delete`).** Open a conversation from its row, open ⋮ and run `control-openhands browser click 'testid=context-menu >> testid=delete-button'`; the dialog reads `Confirm Delete` and `Are you sure you want to delete the "<title>" conversation? This action cannot be undone.` with `cancel-button` and `confirm-button`. `control-openhands browser click 'role=dialog >> testid=cancel-button'` keeps the row. Repeat and `control-openhands browser click 'role=dialog >> testid=confirm-button' --expect-url '/conversations(\?|$)'`; after `browser reload` the row count is `0` and `control-openhands api GET /api/conversations/<id>` is `404`.- **Phone (`F04.phone`).** Run `control-openhands browser viewport phone`, `control-openhands browser goto /`, `control-openhands browser click 'testid=sidebar-mobile-menu-toggle'`, then `control-openhands browser bbox 'testid=sidebar-mobile-drawer >> testid=conversation-panel'` (`insideViewport` `true`, `pageHorizontalOverflow` `false`) and `control-openhands browser screenshot --feature F04.phone --name drawer`. Hover a row in the drawer and click its `testid=ellipsis-button`; `control-openhands browser bbox 'testid=context-menu'` is inside the viewport. `control-openhands browser mouse-click 380 20` closes the menu (`browser count 'testid=context-menu'` `0`) and the drawer (`control-openhands browser wait 'testid=sidebar-mobile-drawer' --state hidden --timeout 5000`; an immediate `visible` still reads `true` during the slide-out). Reopen the drawer and `control-openhands browser click 'testid=sidebar-mobile-drawer >> testid=conversation-card >> has-text=QA_F04 renamed' --expect-url '/conversations/'`; `control-openhands browser wait 'testid=sidebar-mobile-drawer' --state hidden --timeout 5000` succeeds. Return with `control-openhands browser viewport desktop`.+- **A later-page pin (`F04.pin`).** With more than 20 conversations as the pagination precondition already arranges, choose Recent, reload, click `role=button[name="Load more"]`, and pin a conversation from the newly loaded rows (`browser hover` its `conversation-card`, then click its `[data-testid^=conversation-pin-toggle-]`). Read the pinned section, then `control-openhands browser reload` and `control-openhands browser count 'testid=conversation-panel-pinned-section'`. Known failure (reproduced 2026-10-12, [#16399](https://github.com/OpenHands/OpenHands/issues/16399)): the later-page pin is pruned on first-page reload even though the conversation still exists. Tested with 23 empty conversations and one older pin.+- **Fresh-browser grouping (`F04.grouped-folders`).** Run this after the other list checks: select By workspace, record its groups, `control-openhands browser reset`, `control-openhands onboard --skip`, then `control-openhands browser click 'testid=conversation-layouts-toggle'` and `control-openhands browser click 'testid=layout-preset-by-workspace'`. Compare `browser snapshot 'testid=conversation-panel'` with `conversation status <id>`. Known failure (reproduced 2026-10-12, [#17386](https://github.com/OpenHands/OpenHands/issues/17386)): existing conversations move to No workspace although their server working directory still matches the registered workspace. The reset clears browser metadata; do not use it before checks that depend on the original browser's grouping.- **Delete all (`F04.delete-all`).** Run `control-openhands browser click 'testid=conversation-layouts-toggle'`, `control-openhands browser click 'testid=delete-all-conversations'` and `control-openhands browser snapshot 'role=dialog'`: `Delete all conversations`, `Are you sure you want to delete N conversations?`. Cancel (`role=dialog >> testid=cancel-button`) keeps every row. Confirm with `control-openhands browser click 'role=dialog >> testid=confirm-button'` and `control-openhands browser wait 'role=dialog' --state detached --timeout 30000`; `control-openhands api GET '/api/conversations/search?limit=100'` has no items, Delete all is disabled again, and in By date mode the empty state returns. Expected for every conversation; today N and the deletion cover only the loaded pages (22 conversations, 20 loaded: 2 remain).- **Collapsed rail (`F04.collapsed-rail`).** Run `control-openhands browser click 'testid=sidebar-collapse-toggle'`, then `control-openhands browser count 'testid=conversation-panel'` (`0`). Expand with `control-openhands browser click 'aside[data-collapsed]'` (see F02); the count is `1`.- **Cloud picker and start tasks (`F04.cloud-picker`, `F04.start-task-cards`).** Blocked without an OpenHands Cloud backend and git provider. On Cloud, the same `conversation-panel-new-thread-picker` opens `cloud-provider-tabs`, `cloud-repo-search-input` (`Search repositories`) and `launch-repository` rows; start tasks render `testid=start-task-card` linking to `/conversations/task-<id>`.+- **Cleanup.** View → Delete all twice (the second pass removes the rows the first left behind), until `api GET '/api/conversations/search?limit=100'` is empty; re-apply the Recent preset. Fixture folders and the workspace registry vanish with the run.## Gotchas@@ -140,3 +143,11 @@ Preconditions:- Model-written titles vary from run to run; match finished conversations on a word from the prompt (`has-text=pong`), never on the whole title.- Opening a conversation from the sidebar never changes its tags, also right after an automation setup conversation (one tagged `automationsetup: draft`): the other conversation keeps its tags and opens as a normal chat (recipe in F07 `F07.open-by-url`). A conversation that did get the tag loses it with Edit tags (`remove-tag-automationsetup`).- Known issue #17919: Delete all deletes only the loaded pages, and its dialog states that smaller count (`F04.delete-all`).+- Known issue [#18155](https://github.com/OpenHands/OpenHands/issues/18155) (`F04.show-archived-unarchive`): Missing-runtime conversations shown as Archived but never hidden, chipped, or unarchivable. Live prerequisite: A Cloud sandbox or local runtime that becomes MISSING and non-resumable; ordinary local pause remains resumable.+- Known issue [#17497](https://github.com/OpenHands/OpenHands/issues/17497) (`F04.card`): Corrupted conversation data breaks the entire conversation history list instead of handling individual broken conversations. Live prerequisite: The original corrupted persisted conversation or a deterministic corruption fixture after model-switch failure and fork.+- Known issue [#17257](https://github.com/OpenHands/OpenHands/issues/17257) (`F04.start-task-cards`): Cloud provisioning has no in-progress start-task cards because the list hook always returns an empty array. Live prerequisite: An authenticated Cloud backend producing an in-progress sandbox/repository start task.+- Known issue [#16750](https://github.com/OpenHands/OpenHands/issues/16750) (`F04.card`): Conversation list does not load while automation is running. Live prerequisite: A prompt automation or native delegated task that remains running during conversation-list reload; no usable model is configured here.+- Known issue [#15946](https://github.com/OpenHands/OpenHands/issues/15946) (`F04.card`): Show conversations hidden by active/stale agent-server leases. Live prerequisite: A copied lease-bearing persisted conversation and second owner/expired-lease state; never alter another live run's state.+- Known issue [OpenHands/software-agent-sdk#4987](https://github.com/OpenHands/software-agent-sdk/issues/4987) (`F04.card`): Autotitle input has broken user-message separators and emoji-list formatting; verify the resulting visible title with a real provider. Live prerequisite: A working autotitle model and captured title-generation input/output; Canvas only consumes the resulting server title.+- Known issue [OpenHands/software-agent-sdk#4537](https://github.com/OpenHands/software-agent-sdk/issues/4537) (`F04.card`): A native delegated task can freeze the sidebar conversation list and cause request timeouts until it ends. Live prerequisite: A working model with native TaskTool delegation holding the parent state lock during list reload.+- Known issue [OpenHands/software-agent-sdk#4080](https://github.com/OpenHands/software-agent-sdk/issues/4080) (`F04.card`): One unregistered persisted tool observation can make a whole conversation disappear from the list and return404. Live prerequisite: A copied persisted conversation with its original unregistered observation kind and missing tool-module registration.diff --git a/.agents/skills/verify-openhands/references/feature-map/F05-composer.md b/.agents/skills/verify-openhands/references/feature-map/F05-composer.mdindex fdbd026f1e..d4d9f6aff2 100644--- a/.agents/skills/verify-openhands/references/feature-map/F05-composer.md+++ b/.agents/skills/verify-openhands/references/feature-map/F05-composer.md@@ -58,13 +58,13 @@ Preconditions:- Desktop viewport unless a bullet says otherwise. Microphone dictation is blocked in the sandbox browser (no microphone); `+` > Git tools needs a conversation on a git workspace (not driven here).- **Type and send (`F05.send-message`).** On `/conversations/<id>` run `control-openhands browser attr 'testid=submit-button' aria-label` (`Send message`), `control-openhands browser enabled 'testid=submit-button'` (`false`), `control-openhands browser type 'testid=chat-input' 'Reply with only:'`, `control-openhands browser enabled 'testid=submit-button'` (`true`), `control-openhands browser press Shift+Enter`, `control-openhands browser type 'testid=chat-input' 'ok2'` and `control-openhands browser eval "document.querySelector('[data-testid=chat-input]').innerText"`: `Reply with only:\nok2`. Run `control-openhands browser press Enter --selector 'testid=chat-input'`; the field is empty and `control-openhands browser count 'testid=stop-button'` is `1` while the agent runs. Then `control-openhands conversation wait <id> --fresh --timeout 180` (a fast model can finish before a separate `--until running` wait starts, which then times out on `finished`), `control-openhands browser reload` and `control-openhands browser count 'testid=user-message >> has-text=ok2'`: `1`; `control-openhands browser text 'testid=agent-message >> nth=-1'` is `ok2`.-- **Draft survives navigation (`F05.draft-persistence`).** Run `control-openhands browser type 'testid=chat-input' 'QA draft text'` and wait out the 500 ms save: `control-openhands browser eval "new Promise(r=>setTimeout(()=>r(JSON.parse(localStorage.getItem('conversation-state-<id>')).draftMessage),1500))"` returns `QA draft text`. Then `control-openhands browser click 'testid=backend-selector-settings-link' --expect-url settings`, `control-openhands browser back` and the `innerText` eval above: `QA draft text`. `control-openhands browser reload` restores it again. A draft typed just before leaving is flushed, not lost (#17926, fixed by #17968): `control-openhands browser type 'testid=chat-input' ' plus fast'` and at once `control-openhands browser click 'testid=backend-selector-settings-link' --expect-url settings`, `control-openhands browser back`: the eval reads `QA draft text plus fast`. The same holds for an immediate sidebar switch to another conversation and back (that other conversation's composer stays empty). Clear with `control-openhands browser fill 'testid=chat-input' ''` and wait a second before navigating.+- **Draft survives navigation (`F05.draft-persistence`).** Run `control-openhands browser type 'testid=chat-input' 'QA draft text'` and wait out the 500 ms save: `control-openhands browser eval "new Promise(r=>setTimeout(()=>r(JSON.parse(localStorage.getItem('conversation-state-<id>')).draftMessage),1500))"` returns `QA draft text`. Then `control-openhands browser click 'testid=backend-selector-settings-link' --expect-url settings`, `control-openhands browser back` and the `innerText` eval above: `QA draft text`. `control-openhands browser reload` restores it again. A draft typed just before leaving is flushed, not lost: `control-openhands browser type 'testid=chat-input' ' plus fast'` and at once `control-openhands browser click 'testid=backend-selector-settings-link' --expect-url settings`, `control-openhands browser back`: the eval reads `QA draft text plus fast`. The same holds for an immediate sidebar switch to another conversation and back (that other conversation's composer stays empty). Clear with `control-openhands browser fill 'testid=chat-input' ''` and wait a second before navigating.- **Home draft (`F05.draft-persistence`).** Run `control-openhands browser goto /`, `control-openhands browser type 'testid=chat-input' 'QA home draft'`, `control-openhands browser goto /settings/secrets`, `control-openhands browser goto /`, then the `innerText` eval: `QA home draft`; `control-openhands browser storage --session` lists `oh:home-prompt-draft`. The automation setup page's docked composer (`/automations/setup`, F22) keeps no draft and never touches the Home one. With `QA home draft` still in the Home composer, run `control-openhands browser goto /automations`, `control-openhands browser click 'testid=automations-add-automation'`, `control-openhands browser click 'testid=automations-add-automation-create' --expect-url '/automations/setup'` and `control-openhands browser wait 'testid=chat-input'`: the `innerText` eval is empty. `control-openhands browser type 'testid=chat-input' 'Reply with only the word ok2. Do not run any tools.'`, then `control-openhands browser eval "new Promise(r=>setTimeout(()=>r(sessionStorage.getItem('oh:home-prompt-draft')),1500))"` still returns `QA home draft`. `control-openhands browser click 'testid=submit-button' --expect-url '/conversations/'` and `control-openhands browser click 'testid=sidebar-conversations-link' --expect-url '/conversations(\?|$)'`: the Home composer shows `QA home draft`. Clear it (`fill 'testid=chat-input' ''`, wait a second) and repeat the setup send: New Chat is empty and `browser storage --session --values` has no `oh:home-prompt-draft`. Unsent text in the setup composer is lost when the user leaves that page (by design).- **Auto-resize (`F05.composer-resize`).** On `/conversations/<id>` (`control-openhands browser goto /conversations/<id>`; Home draft left `/`) type `line`, then eight times `control-openhands browser press Shift+Enter` and `control-openhands browser type 'testid=chat-input' 'lineN'` (`line2`…`line9`): `control-openhands browser bbox 'testid=chat-input'` grows by 20 px per line, from `height` `20` to `180`; after `control-openhands browser fill 'testid=chat-input' ''` it is `20`. `control-openhands browser count '#resize-grip'` is `1`. `control-openhands browser drag '#resize-grip' --by 0,-200` makes the empty field `220` high and `control-openhands browser drag '#resize-grip' --by 0,150` brings it to `70`; after `control-openhands browser reload` it is `20` again (the dragged height is not kept).- **Tools menu and macros (`F05.plus-menu`).** On `/conversations/<id>` run `control-openhands browser click 'testid=chat-plus-button'` and `control-openhands browser testids 'testid=tools-context-menu'`: `macros-button`, `show-skills-button`, `show-hooks-button`, `show-agent-tools-button`, `add-files-and-images-button` (no `git-tools-button` outside a repository). `control-openhands browser hover 'testid=macros-button'` shows `macros-submenu` with `increase-test-coverage-button`, `fix-readme-button`, `auto-merge-prs-button`, `clean-dependencies-button`. `control-openhands browser click 'testid=fix-readme-button'` closes the menu and fills the composer with `Please look at the README and make the following improvements…`; the `testid=user-message` count is unchanged. Clear with `control-openhands browser fill 'testid=chat-input' ''`. Open the menu again and `control-openhands browser press Escape`: `browser count 'testid=tools-context-menu'` is `0` and `control-openhands browser eval "document.activeElement.dataset.testid"` is `chat-plus-button`.- **Show dialogs (`F05.plus-menu`).** For each of `show-skills-button`, `show-hooks-button`, `show-agent-tools-button`: `control-openhands browser click 'testid=chat-plus-button'`, `control-openhands browser click 'testid=show-hooks-button'` (etc.), `control-openhands browser text 'role=dialog'` (starts `Available Skills`, `Available Hooks`, `Agent Tools & Metadata`), then `control-openhands browser press Escape` and `control-openhands browser count 'role=dialog'`: `0`.- **Agent profile before start (`F05.agent-profile-switch`).** Run `control-openhands browser goto /`, `control-openhands browser click 'testid=chat-plus-button'`, `control-openhands browser hover 'testid=switch-agent-profile-button'` and `control-openhands browser text 'testid=agent-profile-submenu'`: `AVAILABLE PROFILES`, `default` (checked; `chat-input-agent-profile-option-default`), `Manage agent profiles`. `control-openhands browser click 'testid=agent-profile-submenu >> text=Manage agent profiles' --expect-url 'settings/agents'` lands on `/settings/agents`. In a started conversation (`control-openhands browser goto /conversations/<id>`) `control-openhands browser click 'testid=chat-plus-button'` then `control-openhands browser count 'testid=switch-agent-profile-button'` is `0`.-- **Attach a file and an image (`F05.attach-files`).** On `/conversations/<id>` run `control-openhands browser upload 'testid=upload-image-input' "$OH_VERIFY_RUN/evidence/_fixtures/qa-image.png" "$OH_VERIFY_RUN/evidence/_fixtures/qa-note.txt"` and `control-openhands browser screenshot 'testid=interactive-chat-box' --feature F05.attach-files --name attached`: a `qa-note.txt TXT` chip and a thumbnail. `control-openhands browser count 'testid=interactive-chat-box >> alt=qa-image.png'` is `1`. `control-openhands browser click 'role=button[name="Upload as file"]'` turns the image toggle into `Do not upload as file` (`control-openhands browser attr 'role=button[name="Do not upload as file"]' aria-pressed` is `true`); click it again to restore. The remove buttons are named: `control-openhands browser snapshot 'testid=interactive-chat-box'` lists `button "Remove qa-note.txt"` and `button "Remove image"`. Remove the image with `control-openhands browser click 'testid=interactive-chat-box >> role=button[name="Remove image"]'`; the `alt=qa-image.png` count is `0`. Then `control-openhands browser type 'testid=chat-input' 'Reply with only the content of the attached file.'`, `control-openhands browser press Enter --selector 'testid=chat-input'`, wait as in Type and send, and `control-openhands browser text 'testid=user-message >> nth=-1'`: it ends `NEW FILES ADDED: qa-note.txt`. `control-openhands conversation events <id> --kinds MessageEvent --last 3` shows the reply `QA attachment 7731`, and `qa-note.txt` sits in the conversation `workspace` printed by `control-openhands conversation status <id>`. Attachments added on Home before the conversation exists upload after it is created: with no workspace or a Local Repo workspace they land in it (`browser network --filter upload` shows `200`). Expected: the same in New Worktree mode. Known failure (reproduced 2026-10-14, #18421): run `control-openhands fixture git-repo --name qa-repo`, `control-openhands browser goto /`, `control-openhands workspace open qa-repo`, `control-openhands browser click 'testid=workspace-mode-selector'`, `control-openhands browser click 'testid=workspace-mode-selector-option-new_worktree'`, `control-openhands browser click 'testid=chat-plus-button'`, `control-openhands browser upload-via 'testid=add-files-and-images-button' "$OH_VERIFY_RUN/evidence/_fixtures/qa-note.txt"`, type `Reply with only the word received. Do not run any tools.` and `control-openhands browser click 'testid=home-chat-launcher >> testid=submit-button' --expect-url '/conversations/[^/]+$'`: `control-openhands browser toasts --history` shows `HTTP request failed (422 Unprocessable Entity): {"detail":"path must be inside the conversation workspace"}` and `browser network --filter upload` a `422` for `?path=<run>/workspace/qa-repo/qa-note.txt`, while `conversation status <id>` reports the worktree `/tmp/conversation-worktrees/<id>/qa-repo`; the message is sent without the file. The same file attached inside that conversation afterwards uploads to the worktree (`200`). Restore with `testid=workspace-mode-selector-option-local_repo`.+- **Attach a file and an image (`F05.attach-files`).** On `/conversations/<id>` run `control-openhands browser upload 'testid=upload-image-input' "$OH_VERIFY_RUN/evidence/_fixtures/qa-image.png" "$OH_VERIFY_RUN/evidence/_fixtures/qa-note.txt"` and `control-openhands browser screenshot 'testid=interactive-chat-box' --feature F05.attach-files --name attached`: a `qa-note.txt TXT` chip and a thumbnail. `control-openhands browser count 'testid=interactive-chat-box >> alt=qa-image.png'` is `1`. `control-openhands browser click 'role=button[name="Upload as file"]'` turns the image toggle into `Do not upload as file` (`control-openhands browser attr 'role=button[name="Do not upload as file"]' aria-pressed` is `true`); click it again to restore. The remove buttons are named: `control-openhands browser snapshot 'testid=interactive-chat-box'` lists `button "Remove qa-note.txt"` and `button "Remove image"`. Remove the image with `control-openhands browser click 'testid=interactive-chat-box >> role=button[name="Remove image"]'`; the `alt=qa-image.png` count is `0`. Then `control-openhands browser type 'testid=chat-input' 'Reply with only the content of the attached file.'`, `control-openhands browser press Enter --selector 'testid=chat-input'`, wait as in Type and send, and `control-openhands browser text 'testid=user-message >> nth=-1'`: it ends `NEW FILES ADDED: qa-note.txt`. `control-openhands conversation events <id> --kinds MessageEvent --last 3` shows the reply `QA attachment 7731`, and `qa-note.txt` sits in the conversation `workspace` printed by `control-openhands conversation status <id>`. Attachments added on Home before the conversation exists upload after it is created: with no workspace or a Local Repo workspace they land in it (`browser network --filter upload` shows `200`). Expected: the same in New Worktree mode. Known failure (reproduced 2026-10-16, SDK 1.55.0, [#18421](https://github.com/OpenHands/OpenHands/issues/18421)): run `control-openhands fixture git-repo --name qa-repo`, `control-openhands browser goto /`, `control-openhands workspace open qa-repo`, `control-openhands browser click 'testid=workspace-mode-selector'`, `control-openhands browser click 'testid=workspace-mode-selector-option-new_worktree'`, `control-openhands browser click 'testid=chat-plus-button'`, `control-openhands browser upload-via 'testid=add-files-and-images-button' "$OH_VERIFY_RUN/evidence/_fixtures/qa-note.txt"`, type `Reply with only the word received. Do not run any tools.` and `control-openhands browser click 'testid=home-chat-launcher >> testid=submit-button' --expect-url '/conversations/[^/]+$'`: `control-openhands browser toasts --history` shows `HTTP request failed (422 Unprocessable Entity): {"detail":"path must be inside the conversation workspace"}` and `browser network --filter upload` a `422` for `?path=<run>/workspace/qa-repo/qa-note.txt`, while `conversation status <id>` reports the worktree `/tmp/conversation-worktrees/<id>/qa-repo`; the message is sent without the file. The same file attached inside that conversation afterwards uploads to the worktree (`200`). Restore with `testid=workspace-mode-selector-option-local_repo`.```sh known-bug F05.attach-files home-new-worktreecontrol-openhands browser viewport desktopcontrol-openhands fixture git-repo --name qa-repo@@ -86,7 +86,7 @@ Preconditions:control-openhands assert --field recent.0.status --equals 200 --bug-shows 422 -- browser network --filter file/upload # bug```- **Size limit (`F05.attach-size-limit`).** Arrange `head -c 4194304 /dev/zero > "$OH_VERIFY_RUN/evidence/_fixtures/qa-big.bin"` (and `qa-two-a.bin`, `qa-two-b.bin` with `head -c 2097152`). Run `control-openhands browser upload 'testid=upload-image-input' "$OH_VERIFY_RUN/evidence/_fixtures/qa-big.bin"` and `control-openhands browser toasts`: `Error: Files exceeding 3MB are not allowed: qa-big.bin`. Uploading `qa-two-a.bin` and `qa-two-b.bin` together toasts `Error: Total file size would be 4.0MB, exceeding the 3MB limit. Please select fewer or smaller files.` `control-openhands browser count 'testid=interactive-chat-box >> text=qa-big.bin'` is `0`.-- **Image alone (`F05.image-only-send`).** From the home composer: `control-openhands browser goto /`, `control-openhands browser enabled 'testid=submit-button'` (`false`, empty field), `control-openhands browser click 'testid=chat-plus-button'`, `control-openhands browser upload-via 'testid=add-files-and-images-button' "$OH_VERIFY_RUN/evidence/_fixtures/qa-image.png"`, then `control-openhands browser count 'testid=home-chat-launcher >> alt=qa-image.png'` (`1`) and `control-openhands browser enabled 'testid=submit-button'`: `true` while `control-openhands browser eval "document.querySelector('[data-testid=chat-input]').innerText"` is still empty (`control-openhands browser screenshot 'testid=home-chat-launcher' --feature F05.image-only-send --name image-alone-send-enabled`). Run `control-openhands browser click 'testid=submit-button' --expect-url '/conversations/'` and read `<image-id>` from `control-openhands browser url`. After `control-openhands browser wait 'testid=user-message'` and `control-openhands browser wait 'testid=chat-message-sending' --state detached --timeout 5000` (the pending bubble carries `testid=user-message` too, so wait for its `Sending...` to go, not only for a bubble; the wait returns `"state": "detached"` at once when the indicator has already gone, the usual case on a local stack), `control-openhands browser count 'testid=user-message >> testid=image-preview'` is `1`, and `control-openhands browser count 'testid=chat-message-sending'` and `control-openhands browser count 'testid=chat-message-retry'` are `0`. Server side, `control-openhands conversation events <image-id> --kinds MessageEvent --grep 'data:image/png;base64'` matches the one user row (`count` `1`; its `text` is a blank `" "`, the row carries `images: 1`, and the excerpt shows the `"type":"image"` block with `image_urls`), and `control-openhands conversation events <image-id> --kinds MessageEvent` (default `--last`) lists the same row with `images: 1` and no excerpt. `--kinds` filters before `--last` counts, so `--kinds MessageEvent --last 1` is that same row. To check that the model is handed the picture, send the image from Home together with `Which two colours are the bars in this image? Answer with the colour names only. Do not run any tools.` and read `control-openhands conversation events <id> --kinds MessageEvent`. Expect `orange` and `blue` in the agent row and `--kinds ActionEvent` `count` `0` (`fixture image` draws those bars). Known failure (reproduced 2026-10-08, agent-server 1.53.0, the pinned version): on a `deepseek/*` profile the user row still carries `images: 1`, but the agent answers that no image was attached. The model gets text only. The cause is in the SDK, not Canvas: 1.53.0 string-serializes every model id containing `deepseek`, which drops image parts. OpenHands/software-agent-sdk#5460 and #5467 fixed it on 2026-10-08, after the 1.53.0 release. On a stack launched with `--sdk-path <software-agent-sdk main>` the same send answers `Blue and orange.` with no tool call. An image-only send there reaches the model too: its first reasoning describes four blue and orange stripes. On an affected stack, the image-only send leaves the model with an empty task, and deepseek-flash then explores the disk (see Gotchas). Return with `control-openhands browser goto /conversations/<id>`.+- **Image alone (`F05.image-only-send`).** From the home composer: `control-openhands browser goto /`, `control-openhands browser enabled 'testid=submit-button'` (`false`, empty field), `control-openhands browser click 'testid=chat-plus-button'`, `control-openhands browser upload-via 'testid=add-files-and-images-button' "$OH_VERIFY_RUN/evidence/_fixtures/qa-image.png"`, then `control-openhands browser count 'testid=home-chat-launcher >> alt=qa-image.png'` (`1`) and `control-openhands browser enabled 'testid=submit-button'`: `true` while `control-openhands browser eval "document.querySelector('[data-testid=chat-input]').innerText"` is still empty (`control-openhands browser screenshot 'testid=home-chat-launcher' --feature F05.image-only-send --name image-alone-send-enabled`). Run `control-openhands browser click 'testid=submit-button' --expect-url '/conversations/'` and read `<image-id>` from `control-openhands browser url`. After `control-openhands browser wait 'testid=user-message'` and `control-openhands browser wait 'testid=chat-message-sending' --state detached --timeout 5000` (the pending bubble carries `testid=user-message` too, so wait for its `Sending...` to go, not only for a bubble; the wait returns `"state": "detached"` at once when the indicator has already gone, the usual case on a local stack), `control-openhands browser count 'testid=user-message >> testid=image-preview'` is `1`, and `control-openhands browser count 'testid=chat-message-sending'` and `control-openhands browser count 'testid=chat-message-retry'` are `0`. Server side, `control-openhands conversation events <image-id> --kinds MessageEvent --grep 'data:image/png;base64'` matches the one user row (`count` `1`; its `text` is a blank `" "`, the row carries `images: 1`, and the excerpt shows the `"type":"image"` block with `image_urls`), and `control-openhands conversation events <image-id> --kinds MessageEvent` (default `--last`) lists the same row with `images: 1` and no excerpt. `--kinds` filters before `--last` counts, so `--kinds MessageEvent --last 1` is that same row. To check that the model is handed the picture, send the image from Home together with `Which two colours are the bars in this image? Answer with the colour names only. Do not run any tools.` and read `control-openhands conversation events <id> --kinds MessageEvent`. Expect `orange` and `blue` in the agent row and `--kinds ActionEvent` `count` `0` (`fixture image` draws those bars).- **Paste text (`F05.paste`).** Run `control-openhands browser type 'testid=chat-input' 'QA paste text'`, `control-openhands browser press Control+a`, `control-openhands browser press Control+c`, `control-openhands browser press Delete`, `control-openhands browser press Control+v`, then `control-openhands browser eval "document.querySelector('[data-testid=chat-input]').innerHTML"`: `QA paste text`, no markup. On a cleared field `control-openhands browser paste 'testid=chat-input' --file "$OH_VERIFY_RUN/evidence/_fixtures/qa-image.png"` attaches the image (`control-openhands browser count 'testid=interactive-chat-box >> alt=qa-image.png'` is `1`, with an `Upload as file` toggle) and `control-openhands browser paste 'testid=chat-input' --text 'QA pasted plain'` inserts plain text (`innerHTML` is `QA pasted plain`). `control-openhands browser drop-files 'testid=chat-input' "$OH_VERIFY_RUN/evidence/_fixtures/qa-note.txt" --stage over` makes `control-openhands browser text 'testid=interactive-chat-box'` start `Drop your files here`; the same command without `--stage` drops it: a `qa-note.txt` `TXT` chip appears. Remove the chip with `control-openhands browser click 'testid=interactive-chat-box >> role=button[name="Remove qa-note.txt"]'` and the image as in Attach; clear the field.- **Dictation failure (`F05.dictation`).** Run `control-openhands browser click 'testid=chat-dictation-button' --observe '[role=status]' --observe-ms 3000`: the observed toasts include `Dictation failed. Check microphone access and your voice input settings.` and `control-openhands browser attr 'testid=chat-dictation-button' aria-label` is back to `Start dictation`. Real dictation is blocked without a microphone.- **Switch LLM profile (`F05.llm-profile-picker`).** Run `control-openhands browser click 'testid=chat-input-llm-profile'` and `control-openhands browser text 'testid=chat-input-llm-profile-popover'`: `AVAILABLE PROFILES`, `deepseek-flash` / `deepseek/deepseek-flash`, `deepseek-pro` / `deepseek/deepseek-v4-pro`, `LLM Profiles`. Run `control-openhands browser click 'testid=chat-input-llm-profile-option-deepseek-pro'`, `control-openhands browser wait-text 'Switched to profile' --timeout 15000`, `control-openhands browser text 'testid=chat-input-llm-profile'` (`deepseek-pro`) and `control-openhands conversation status <id>` (`model` is `deepseek/deepseek-v4-pro`). Send `Reply with only: pro-ok` as in Type and send; the reply is `pro-ok`. After `control-openhands browser reload` the pill still reads `deepseek-pro`. Switch back with `control-openhands browser click 'testid=chat-input-llm-profile'` and `control-openhands browser click 'testid=chat-input-llm-profile-option-deepseek-flash'`.@@ -105,12 +105,15 @@ Preconditions:- **Build button (`F05.plan-preview`).** With the newer plan card from `/plan Plan writing bye.txt containing bye. One step only.` (above) shown and the agent idle, `control-openhands browser click 'testid=plan-preview-build-button >> nth=-1'` does the same as the shortcut: placeholder `What do you want to build?`, `control-openhands conversation wait <id> --fresh --timeout 240`, the same last user message, and `bye.txt` (content `bye`) in the workspace.- **Phone (`F05.phone`).** On `/conversations/<id>` run `control-openhands browser viewport phone`, `control-openhands browser bbox 'testid=interactive-chat-box'` (`insideViewport` `true`, `pageHorizontalOverflow` `false`) and `control-openhands browser screenshot --feature F05.phone --name composer` (`+`, model pill, ring, mic and Send on one row). Return with `control-openhands browser viewport desktop`.+- **Home file re-pick (`F05.attach-files`).** As a separate check after the conversation recipes, run `control-openhands browser goto /`, upload the file fixture with `control-openhands browser upload 'testid=upload-image-input' "$OH_VERIFY_RUN/evidence/_fixtures/qa-note.txt"`, remove it with `control-openhands browser click 'role=button[name="Remove qa-note.txt"]'`, upload that same path again, then `control-openhands browser count 'role=button[name="Remove qa-note.txt"]'`. Expected: `1`. Known failure (reproduced 2026-10-12, [#17471](https://github.com/OpenHands/OpenHands/issues/17471)): the second pick is ignored and the count is `0`. End with `browser reload` to clear the file input.+- **Unmatched failed-conversation profile (`F05.profile-identity`).** Separate disposable check: record the original active profile with `control-openhands llm show`, then create an empty Local conversation with a new dummy profile pointing to closed loopback `http://127.0.0.1:9` (`llm set --no-validate`, as F06's dummy-profile precondition does); give it model `openai/gpt-4o`. After `browser reset` and `onboard --skip`, reopen it and send `Reply with only QA. Do not run any tools.`. Change the same saved profile to `openai/gpt-4o-mini` with `llm set --no-validate`, then reload and read `browser text 'testid=chat-input-llm-profile'` and `conversation status <id>`. Known failure (reproduced 2026-10-12, [#16263](https://github.com/OpenHands/OpenHands/issues/16263)): once the loopback turn errors, the pill reads Select a model despite a loaded active profile and the persisted conversation model. This check covers the missing browser stamp and unmatched-model case, not successful provider output. Reactivate the original profile with `control-openhands api POST /api/profiles/<original-profile>/activate --write`, then delete the disposable conversation and profile afterwards.+## Gotchas- Always export your own `OH_VERIFY_RUN`. `launch --new` repoints the shared `current` symlink, and another agent relying on `current` drove and then stopped this family's stack mid-run. If the URL jumps to a page you never opened or `fetch failed` appears, run `control-openhands status` before blaming the product.- `conversation wait <id>` right after sending returns at once with the previous `finished` status. Use `conversation wait <id> --fresh`; `--until running` then a terminal wait only works for turns longer than the poll interval (a `Reply with only` turn on deepseek-flash can finish first, and `--until running` then times out).- The slash menu stays open while you type a command; press Escape before Enter or Enter selects the highlighted option instead of sending.-- Conversation drafts are saved 500 ms after the last keystroke and flushed on navigation and unmount (#17926, fixed by #17968), so text typed just before leaving comes back.+- Conversation drafts are saved 500 ms after the last keystroke and flushed on navigation and unmount, so text typed just before leaving comes back.- Escape closes the `+` tools menu (`browser count 'testid=tools-context-menu'` is `0`) and returns focus to `chat-plus-button`, whether focus was on the trigger or on a menu item. A menu left open covers attachments in screenshots. Escape also closes the narrow-width "More input actions" overflow menu (`browser count 'testid=chat-input-overflow-menu'` is `0`, the trigger's `aria-expanded` is `false`) and returns focus to the trigger, whether focus was on the trigger, on `overflow-model-button` or on a profile row of the open Model submenu (#18174).- After a client-side switch between conversations (a sidebar row, or a user-message branch in `F06.branch-from-here`) Send keeps the previous conversation's state (Known failure, [#18202](https://github.com/OpenHands/OpenHands/issues/18202); [draft fix PR #18203](https://github.com/OpenHands/OpenHands/pull/18203)). Switching from an empty composer to one with a saved draft restores the text next to a disabled Send; this is a race (Send came up enabled in 1 of 7 tries), so one enabled reading does not show the bug is gone. Switching from a composer with text to an empty one leaves Send enabled, and clicking it does nothing. Typing or `browser reload` fixes it, and Enter in the field sends the restored text anyway. Read `browser enabled 'testid=submit-button'` after a reload, or after typing.- Expected: a prompt that a suggestion chip (`testid=chat-suggestions`) or a `+` > **Macros** item writes into the composer enables Send at once. Known failure (reproduced 2026-10-09, [#18202](https://github.com/OpenHands/OpenHands/issues/18202)): on a new No workspace conversation, `browser click 'testid=chat-suggestions >> text=Fix README'` (or `browser click 'testid=chat-plus-button'`, `browser hover 'testid=macros-button'`, `browser click 'testid=fix-readme-button'`) fills the field, but `browser enabled 'testid=submit-button'` is `false` (5 of 5 tries at desktop; a race at phone) until a character is typed. Both paths call `setMessageToSend`, which writes the text without an input event.@@ -126,3 +129,29 @@ Preconditions:- Opening the context popover logs a 404 for `/api/llm/balance` on the local backend in `browser errors --app-only`; it does not break the popover.- `--observe` takes the verbs' selector syntax: `testid=` and plain CSS segments are watched by an in-page MutationObserver (`observedBy` `mutation`), while `role=`, `text=` and other engines are polled every 20 ms (`poll-20ms`), which misses a state shorter than that; prefer a `testid=` or CSS form for anything brief.- At `narrow` (320 px) the overflow menu's Model submenu opens to the right of the menu and runs off screen: each profile row starts at x about 270 with width 210 (`browser bbox 'testid=chat-input-llm-profile-option-deepseek-flash'`, `insideViewport` `false`), so both names read `deepse…`. Pick profiles by test id: the click lands on the strip of about 50 px that stays on screen. Known issue: #18063 (filed from the follow-up noted in #18015; the submenu also runs past the bottom edge, so the LLM Profiles link cannot be clicked at `narrow`) (`F05.overflow-menu`).+- Known issue [#17788](https://github.com/OpenHands/OpenHands/issues/17788) (`F05.attach-files`): Cloud attachment upload rejects valid runtime data when cached fields are empty strings. Live prerequisite: A Cloud sandbox whose cached runtime URL/key are blank while the fresh runtime lookup is already ready.+- Known issue [#17621](https://github.com/OpenHands/OpenHands/issues/17621) (`F05.context-window-meter`): Compact context always fails after exactly 60s — client-side timeout aborts a condensation the server then completes. Live prerequisite: A real summarizing model and conversation whose condensation takes longer than 60 seconds.+- Known issue [#17476](https://github.com/OpenHands/OpenHands/issues/17476) (`F05.context-window-meter`): Planning socket stats overwrite the main conversation's context meter. Live prerequisite: A real planning sub-conversation and main conversation with distinct nonzero usage stats.+- Known issue [#17429](https://github.com/OpenHands/OpenHands/issues/17429) (`F05.slash-btw`): 500 error with /btw. Live prerequisite: The exact pending tool/streaming state that triggers the generator-pickle error, plus a working model for the side question.+- Known issue [#17271](https://github.com/OpenHands/OpenHands/issues/17271) (`F05.context-window-meter`): New chat retains context/state from the previous conversation. Live prerequisite: A completed conversation with nonzero token usage before New Chat; a dummy-key run supplies no genuine context usage.+- Known issue [#17263](https://github.com/OpenHands/OpenHands/issues/17263) (`F05.context-window-meter`): Context meter stays pinned after a successful compaction once the condenser has its own usage entry. Live prerequisite: A real successful compaction whose condenser usage remains larger than the default agent usage.+- Known issue [#16851](https://github.com/OpenHands/OpenHands/issues/16851) (`F05.profile-identity`): wrong model is displayed when loading existing conversation. Live prerequisite: The original conversation/model-switch history and usage record showing Opus with a Fable selector; a dummy-key conversation has no usage evidence.+- Known issue [#16835](https://github.com/OpenHands/OpenHands/issues/16835) (`F05.context-window-meter`): Agent Canvas sidebar shows "Cache Hit: 0" while backend correctly records nonzero cache_read_tokens. Live prerequisite: A working prompt-caching model/proxy and a real conversation reporting nonzero cache_read_tokens.+- Known issue [#16430](https://github.com/OpenHands/OpenHands/issues/16430) (`F05.attach-files`): File upload does not have an indication of progress. Live prerequisite: A remote or throttled real upload that stays in flight long enough to inspect progress and composer state.+- Known issue [#16339](https://github.com/OpenHands/OpenHands/issues/16339) (`F05.attach-files`): Agent not informed of file location for files uploaded via canvas in cloud. Live prerequisite: An authenticated Cloud upload sandbox and working model to inspect the delivered filename/path notification and read the uploaded file.+- Known issue [#15782](https://github.com/OpenHands/OpenHands/issues/15782) (`F05.draft-persistence`): Prompt disappears when a conversation is stopped. Live prerequisite: Clarify whether prompt means unsent composer text or the already-sent history message, and capture that state before sidebar Stop.+- Known issue [OpenHands/software-agent-sdk#5639](https://github.com/OpenHands/software-agent-sdk/issues/5639) (`F05.llm-profile-picker`): A profile switched before the first message can be absent from token/cost statistics after its successful turn. Live prerequisite: A working model, a profile switch before the first run, and nonzero usage/cost after that turn.+- Known issue [OpenHands/software-agent-sdk#5627](https://github.com/OpenHands/software-agent-sdk/issues/5627) (`F05.slash-goal`): After backend restart the goal chip can remain Running while the conversation is paused, and Stop does not clear it. Live prerequisite: A working model goal loop active during graceful backend restart, then goal-status and resume readback.+- Known issue [OpenHands/software-agent-sdk#5523](https://github.com/OpenHands/software-agent-sdk/issues/5523) (`F05.image-only-send`): DeepSeek V4.1 Flash images can be sent to another fallback profile because its vision capability is not detected. Live prerequisite: A working DeepSeek V4.1 Flash vision profile and another fallback profile, with an image-bearing turn.+- Known issue [OpenHands/software-agent-sdk#5492](https://github.com/OpenHands/software-agent-sdk/issues/5492) (`F05.slash-goal`): A goal can stop as Complete when its judge replies with complete as the string false. Live prerequisite: A real goal judge response containing complete as the string false rather than a JSON boolean.+- Known issue [OpenHands/software-agent-sdk#5330](https://github.com/OpenHands/software-agent-sdk/issues/5330) (`F05.slash-btw`): A /btw side question can prevent reuse of the normal conversation prompt cache. Live prerequisite: A working caching provider and comparable ordinary-message versus /btw prompt/cache records.+- Known issue [OpenHands/software-agent-sdk#4498](https://github.com/OpenHands/software-agent-sdk/issues/4498) (`F05.context-window-meter`): Compaction can leave an orphaned tool call/result in the live history and make the next provider request fail. Live prerequisite: A real condensation splitting an action/result pair, followed by the next provider request.+- Known issue [OpenHands/software-agent-sdk#4382](https://github.com/OpenHands/software-agent-sdk/issues/4382) (`F05.context-window-meter`): ACP cost can miss cache/thought usage, silently become zero, or inflate after stalled/reset cost updates. Live prerequisite: An authenticated ACP provider emitting cache/thought token usage or reset/stalled cumulative cost updates.++- Known issue [#17239](https://github.com/OpenHands/OpenHands/issues/17239) (`F05.llm-profile-picker`): The reported model/profile picker can open above the viewport with unreachable rows. A frozen-target Local check with eight LLM profiles passed on Home at 1512x720, 1440x1000, 390x844 and 320x700, and in a genuine conversation at 1512x720 and 390x844: internal scrolling and selecting the first/last profile worked. The Cloud screenshot and real ACP multi-model picker remain unverified; they need an approved Cloud session or ACP account/adapter with a multi-row model catalogue. Local profile geometry is not proof of those branches. Run `backfill-18309-settings-worker`, Agent Server/SDK 1.54.0; evidence includes `F05.llm-profile-picker/17239-conversation-1512x720.png` and `17239-conversation-phone.png`.++- Known issue [OpenHands/software-agent-sdk#5824](https://github.com/OpenHands/software-agent-sdk/issues/5824) (`F05.slash-goal`): resuming after a failed judge audit can show iteration 2 of a maximum of 1 and run beyond the selected goal cap. Live prerequisite: an agent/judge run whose audit fails, followed by a successful incomplete judge verdict after Resume; a controller-only test does not establish the Canvas banner outcome.++- Known issue [OpenHands/software-agent-sdk#5889](https://github.com/OpenHands/software-agent-sdk/issues/5889) (`F05.attach-files`): a same-name attachment upload that fails part-way can replace the previous workspace file with partial content. Live prerequisite: an owned isolated backend with a controlled mid-upload filesystem failure (the original Linux file-size limit or an equivalent bounded child-process limit), a pre-existing target and a real Canvas overwrite. Never change resource limits on a shared runtime; a normal successful upload does not verify failure atomicity.++- Known issue [OpenHands/software-agent-sdk#5820](https://github.com/OpenHands/software-agent-sdk/issues/5820) (`F05.context-window-meter`): manual compaction on a null/no-op condenser can return500. Verification is blocked by genuine populated usage/context metrics: idle and refused-loopback runs leave `usage=null`, so neither composer nor Usage panel offers Compact context. A separate direct endpoint check returned500 but is not a Canvas reproduction. Use an existing real conversation with usage and no condenser, or an authorized successful provider run; never fabricate stats to expose the action.diff --git a/.agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md b/.agents/skills/verify-openhands/references/feature-map/F06-agent-activity.mdindex 3061edb486..c9edafe601 100644--- a/.agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md+++ b/.agents/skills/verify-openhands/references/feature-map/F06-agent-activity.md@@ -90,7 +90,7 @@ Preconditions:- **Reload mid-run (`F06.reload-mid-run`).** Run `control-openhands conversation start --prompt "Run exactly one terminal command: sleep 15 && echo qa-reloaded. Then reply with only its output."`, wait for `testid=live-activity-chip`, then `control-openhands browser reload`. The composer still reads `Running` and the chip is back. After `control-openhands conversation wait <id> --timeout 120`: `testid=agent-message` is `qa-reloaded`, `count 'testid=user-message'` is `1`, `control-openhands browser eval "document.querySelectorAll('[data-testid=chat-scroll-container] [data-testid=generic-event-message-title]').length"` is `1`.- **Older history (`F06.load-older-history`).** Run `control-openhands conversation start --prompt "Run these 16 terminal commands strictly one per tool call, sequentially (never combine them): echo qa1, echo qa2, echo qa3, echo qa4, echo qa5, echo qa6, echo qa7, echo qa8, echo qa9, echo qa10, echo qa11, echo qa12, echo qa13, echo qa14, echo qa15, echo qa16. Then reply DONE." --wait --timeout 400` (note `<history-id>`); `control-openhands conversation events <history-id> --last 500` reports a count above 50. Run `control-openhands browser network --clear`, `control-openhands browser reload`, then `control-openhands browser network --last 300`: two `GET /api/conversations/<history-id>/events/search` requests (newest page, then the older page). `control-openhands browser text 'testid=chat-scroll-container'` starts with the prompt and shows `16 actions completed` and `DONE`. The other trigger is the user scrolling to the top, driven without a model on the dummy profile from Preconditions (`qa-zeta` active; in a keyed run create and activate it with that same `llm set` command for this bullet, then re-run `control-openhands llm preset deepseek` and remove it with `control-openhands api DELETE /api/profiles/qa-zeta --write`, so later families still count two profiles; arrange, not proof). Run `control-openhands conversation start --prompt "qa older 1"` (no `--wait`; note `<older-id>`), then `for N in $(seq 2 30); do control-openhands conversation send <older-id> --prompt "qa older $N"; done`: every send is accepted while the agent sits in error, about 1.5 s apiece (42 s for the 29 here), and each refused message adds five events (a `MessageEvent`, three `ConversationStateUpdateEvent`s, a `ConversationErrorEvent`), so the conversation ends with about 150 events (152 here). `control-openhands conversation events <older-id> --last 500` reports `"total": 152`, `"pages": 2`, `"more": false` (both server pages read, see Gotchas); its first row is the `SystemPromptEvent` and the first `MessageEvent` is `qa older 1`, while `--last 25` reports `"pages": 1`, `"more": true` and starts at `qa older 26`. Run `control-openhands browser network --clear`, `control-openhands browser reload`, `control-openhands browser wait 'testid=chat-input'`, then `control-openhands browser network --filter 'events/search' --last 10`: two requests, `?limit=50&sort_order=TIMESTAMP_DESC` and one with `timestamp__lt=<ts of qa older 21>`, because the newest 50 events hold only ten bubbles, shorter than the viewport, so the automatic fetch ran once and then stopped. `control-openhands browser count 'testid=user-message'` is `20`, `control-openhands browser text 'testid=chat-scroll-container'` starts `qa older 11` and holds none of `qa older 1`…`qa older 10`, and `control-openhands browser bbox 'testid=chat-scroll-container'` has `scrollHeight` `1576` against `clientHeight` `814`: taller than the viewport, so the third page waits for the user. Run `control-openhands browser scroll 'testid=chat-scroll-container' --by -100000`: `control-openhands browser network --filter 'events/search' --last 5` gains a request with `timestamp__lt=<ts of qa older 11>`, `control-openhands browser count 'testid=user-message'` is `29` and the text starts `qa older 2` (that page held messages 2–10 plus the tail of the first message's events; not seen in two runs: if the first refusal ever logs six events instead of seven, the arithmetic gives count `30`, text `qa older 1` and a last page of only the `SystemPromptEvent`), and the view stays on `qa older 11`: `control-openhands browser eval "document.querySelector('[data-testid=chat-scroll-container]').scrollTop"` is `684`, the height of the prepended rows. Scroll to the top again the same way: a fourth request (`timestamp__lt=` the oldest loaded event), the count is `30`, the text starts `qa older 1` and `scrollTop` is `76` (`control-openhands browser screenshot 'testid=chat-interface' --feature F06.load-older-history --name scroll-top-loaded-qa-older-1`). A third scroll adds no request: the short page ended the paging. `control-openhands browser count 'testid=loading-older-events'` is `0` after each scroll because the `Fetching older messages…` row lasts a few tens of milliseconds on a local stack; to see it, scroll with `control-openhands browser scroll 'testid=chat-scroll-container' --by -100000 --observe 'testid=loading-older-events' --observe-ms 1500` instead (both scrolls were driven this way here, with the same counts), whose `observed` goes `<absent>` at `0` ms, `Fetching older messages…` at about 20 ms and `<absent>` again within about 40 to 120 ms (`116` ms on the first scroll, `39` on the second; `observedBy` `mutation`); the `timestamp__lt=` network rows remain the durable proof. Return to the 16-command conversation for the next bullet: `control-openhands browser goto /conversations/<history-id>` and `control-openhands browser wait 'testid=chat-input'`.- **Scroll to bottom (`F06.scroll-to-bottom`).** On the 16-command conversation (`<history-id>`; in a model-free run this bullet is blocked with the automatic-fetch half, prerequisite `DEEPSEEK_API_KEY`), `control-openhands browser click 'testid=event-group-toggle'` makes the transcript taller than the viewport while the view stays where it was. Run `control-openhands browser scroll 'testid=chat-scroll-container' --by 2000` first (the button only reacts to scroll events, so a page that is already at the top never shows it): `count 'testid=scroll-to-bottom'` is `0`. Run `control-openhands browser scroll 'testid=chat-scroll-container' --by -2000`; the count is `1`. `control-openhands browser click 'testid=scroll-to-bottom'`, then the count is `0` and `control-openhands browser eval "(()=>{const e=document.querySelector('[data-testid=chat-scroll-container]');return Math.round(e.scrollHeight-e.scrollTop-e.clientHeight)})()"` is `0`.-- **Confirmation mode on (`F06.confirmation-mode`).** Arrange through the settings UI, in this order (the Security Analyzer field appears only once Confirmation Mode is on): `control-openhands browser goto /settings/verification`, `control-openhands browser click 'testid=verification-settings-screen >> text=Confirmation Mode'`, `control-openhands browser click 'testid=sdk-section-all-toggle'`, `control-openhands browser click 'testid=verification-settings-screen >> role=button[name="Show suggestions"]'`, `control-openhands browser click 'role=listbox >> role=option[name="None"]'`, `control-openhands browser value 'testid=verification-settings-screen >> role=combobox[name="Security Analyzer"]'` (`None`), `control-openhands browser click 'testid=verification-settings-screen >> testid=save-button'`; `control-openhands api GET /api/settings` shows `"confirmation_mode": true` and `"security_analyzer": "none"` (None means every action is confirmed; the default LLM analyzer only stops HIGH-risk actions, so `echo` would run unprompted). Run `control-openhands conversation start --prompt "Run the terminal command 'echo qa-confirm-ok' and reply with its output." --wait --timeout 180`: it returns `"status": "waiting_for_confirmation"`. `control-openhands browser text 'testid=chat-scroll-container'` ends `Do you want to continue with this action? Cancel Ctrl+⇧+⌫ Continue Ctrl+↩` on Linux (`Cancel ⇧⌘⌫ Continue ⌘↩` on macOS, #18127), `control-openhands browser tooltip 'testid=action-confirm-button'` is `Confirm the requested action`, and a lock icon sits above the composer (screenshot). The tooltip stays open over the Cancel button, so move the pointer first: `control-openhands browser hover 'testid=chat-input'`. Run `control-openhands browser click 'testid=action-reject-button'`: status `idle`, the buttons are gone, `conversation events <id> --last 6` ends with `UserRejectObservation`, and the action's row reads `Rejected` (`control-openhands browser count 'testid=rejected-indicator'` is `1`) with no spinner. Send `Run the terminal command 'echo qa-confirm-two' and reply with its output.` through `testid=chat-input` / `testid=submit-button`, `control-openhands browser wait 'testid=action-confirm-button' --timeout 20000`, `control-openhands browser click 'testid=action-confirm-button'`; about 15 s later the status is `finished` and the events contain the observation `qa-confirm-two`. A rejection inside a group resolves it too (#17920, fixed by #17995): send `Run the terminal command 'echo qa-g1'. After it finishes, run the terminal command 'echo qa-g2' as a separate tool call. Then reply DONE.`, continue the first action (`browser wait 'testid=action-confirm-button'`, then `control-openhands browser press Control+Enter`, which is also the next bullet's shortcut) and cancel the second (`browser wait 'testid=action-confirm-button' --timeout 60000`, `control-openhands browser press Control+Shift+Backspace`): status `idle`, `control-openhands browser text 'testid=event-group >> nth=-1'` ends `2 actions completed`, `control-openhands browser count 'testid=event-group >> nth=-1 >> .animate-spin'` is `0`, the same after `browser reload`, and the expanded group (`control-openhands browser click 'testid=event-group >> nth=-1 >> button >> nth=0'`) shows `Rejected` on the `qa-g2` row (`browser screenshot 'testid=chat-interface' --feature F06.confirmation-mode --name group-after-reload`). Each button keeps its label and hint on one line: while the prompt is waiting, `control-openhands browser eval "[...document.querySelectorAll('[data-testid=action-reject-button],[data-testid=action-confirm-button]')].map(b=>b.offsetHeight+'/'+b.scrollHeight).join(' ')"` is `26/26 26/26` at desktop, also with the side panel open (`testid=right-panel-toggle`). At phone width (`browser viewport phone`) the shortcut hints are hidden (no hardware keyboard), the buttons read `Cancel` and `Continue`, and `browser text 'testid=chat-scroll-container'` ends `Do you want to continue with this action? Cancel Continue` (screenshot the awaiting prompt at phone width).+- **Confirmation mode on (`F06.confirmation-mode`).** Arrange through the settings UI, in this order (the Security Analyzer field appears only once Confirmation Mode is on): `control-openhands browser goto /settings/verification`, `control-openhands browser click 'testid=verification-settings-screen >> text=Confirmation Mode'`, `control-openhands browser click 'testid=sdk-section-all-toggle'`, `control-openhands browser click 'testid=verification-settings-screen >> role=button[name="Show suggestions"]'`, `control-openhands browser click 'role=listbox >> role=option[name="None"]'`, `control-openhands browser value 'testid=verification-settings-screen >> role=combobox[name="Security Analyzer"]'` (`None`), `control-openhands browser click 'testid=verification-settings-screen >> testid=save-button'`; `control-openhands api GET /api/settings` shows `"confirmation_mode": true` and `"security_analyzer": "none"` (None means every action is confirmed; the default LLM analyzer only stops HIGH-risk actions, so `echo` would run unprompted). Run `control-openhands conversation start --prompt "Run the terminal command 'echo qa-confirm-ok' and reply with its output." --wait --timeout 180`: it returns `"status": "waiting_for_confirmation"`. `control-openhands browser text 'testid=chat-scroll-container'` ends `Do you want to continue with this action? Cancel Ctrl+⇧+⌫ Continue Ctrl+↩` on Linux (`Cancel ⇧⌘⌫ Continue ⌘↩` on macOS, #18127), `control-openhands browser tooltip 'testid=action-confirm-button'` is `Confirm the requested action`, and a lock icon sits above the composer (screenshot). The tooltip stays open over the Cancel button, so move the pointer first: `control-openhands browser hover 'testid=chat-input'`. Run `control-openhands browser click 'testid=action-reject-button'`: status `idle`, the buttons are gone, `conversation events <id> --last 6` ends with `UserRejectObservation`, and the action's row reads `Rejected` (`control-openhands browser count 'testid=rejected-indicator'` is `1`) with no spinner. Send `Run the terminal command 'echo qa-confirm-two' and reply with its output.` through `testid=chat-input` / `testid=submit-button`, `control-openhands browser wait 'testid=action-confirm-button' --timeout 20000`, `control-openhands browser click 'testid=action-confirm-button'`; about 15 s later the status is `finished` and the events contain the observation `qa-confirm-two`. A rejection inside a group resolves it too: send `Run the terminal command 'echo qa-g1'. After it finishes, run the terminal command 'echo qa-g2' as a separate tool call. Then reply DONE.`, continue the first action (`browser wait 'testid=action-confirm-button'`, then `control-openhands browser press Control+Enter`, which is also the next bullet's shortcut) and cancel the second (`browser wait 'testid=action-confirm-button' --timeout 60000`, `control-openhands browser press Control+Shift+Backspace`): status `idle`, `control-openhands browser text 'testid=event-group >> nth=-1'` ends `2 actions completed`, `control-openhands browser count 'testid=event-group >> nth=-1 >> .animate-spin'` is `0`, the same after `browser reload`, and the expanded group (`control-openhands browser click 'testid=event-group >> nth=-1 >> button >> nth=0'`) shows `Rejected` on the `qa-g2` row (`browser screenshot 'testid=chat-interface' --feature F06.confirmation-mode --name group-after-reload`). Each button keeps its label and hint on one line: while the prompt is waiting, `control-openhands browser eval "[...document.querySelectorAll('[data-testid=action-reject-button],[data-testid=action-confirm-button]')].map(b=>b.offsetHeight+'/'+b.scrollHeight).join(' ')"` is `26/26 26/26` at desktop, also with the side panel open (`testid=right-panel-toggle`). At phone width (`browser viewport phone`) the shortcut hints are hidden (no hardware keyboard), the buttons read `Cancel` and `Continue`, and `browser text 'testid=chat-scroll-container'` ends `Do you want to continue with this action? Cancel Continue` (screenshot the awaiting prompt at phone width).- **Confirmation shortcuts (`F06.confirmation-shortcuts`).** The previous bullet resolved its action, so arrange a new one in the same conversation: send `Run the terminal command 'echo qa-confirm-three' and reply with its output.` through `testid=chat-input` / `testid=submit-button` and `control-openhands browser wait 'testid=action-confirm-button' --timeout 20000`. `control-openhands browser text 'testid=action-confirm-button'` is `Continue Ctrl+↩` off macOS. Ctrl works off macOS (#17928, fixed by #18127): the previous bullet's `Control+Enter` continued and `Control+Shift+Backspace` cancelled. `control-openhands browser press Meta+Enter` continues too: `control-openhands conversation wait <id> --fresh --timeout 120` ends `finished` and `control-openhands conversation events <id> --kinds ObservationEvent` contains `qa-confirm-three`. Restore afterwards: `browser goto /settings/verification`, click `testid=sdk-section-all-toggle`, `Show suggestions`, `role=listbox >> role=option[name="LLM"]`, then `testid=verification-settings-screen >> text=Confirmation Mode`, then `save-button`; `api GET /api/settings` shows `"confirmation_mode": false`, `"security_analyzer": "llm"`.- **Branch from here (`F06.branch-from-here`).** Open a conversation with at least two exchanges (the OK/PONG one above): `control-openhands browser goto /conversations/<id>` with its id (the previous bullet left `/settings/verification`). Agent message: `control-openhands browser hover 'testid=agent-message >> nth=0'`, `control-openhands browser click 'testid=agent-message >> nth=0 >> role=button[name="Branch from here"]' --expect-url '/conversations/(?!<id>)'`; after `control-openhands browser wait 'testid=user-message'` (the history mounts a moment after the URL changes, so an immediate count reads `0`) the new conversation has `count 'testid=user-message'` `1`, the agent message `OK`, and `browser eval "document.title"` contains `(branch)`. User message (edit): back on the OK/PONG conversation (`control-openhands browser goto /conversations/<id>`; the branch above has one user message), `control-openhands browser hover 'testid=user-message >> nth=1'` and `control-openhands browser click 'testid=user-message >> nth=1 >> role=button[name="Branch from here"]' --observe '[data-testid=chat-input]' --observe-ms 4000`: `observed` shows the composer change from empty to `Reply with only the word PONG.` and stay. The new conversation (`<branch-id>` from `browser url`) omits that message, `control-openhands browser text 'testid=chat-input'` is `Reply with only the word PONG.`, and nothing was sent: `control-openhands conversation events <branch-id> --kinds MessageEvent` lists only the copied first exchange (count `2`: the long message that starts `Reply with only the word OK.`, then `OK`; no `PONG`). `control-openhands browser enabled 'testid=submit-button'` should be `true` with the prefilled text; today it is `false` until the text is edited or the page is reloaded (see Gotchas). Send it with `control-openhands browser press Enter --selector 'testid=chat-input'` (Enter sends whatever the Send button's state is) and `control-openhands conversation wait <branch-id> --fresh --timeout 120`: expected `finished` with a reply; today it ends `error` (Agent Server bug, see Gotchas).- **Images (`F06.image-attachments`).** Run `control-openhands fixture image --name qa-f06-img` (prints the PNG path), `control-openhands browser upload 'testid=upload-image-input' <png path>` (a thumbnail appears in the composer), fill `testid=chat-input` with `Reply with only the word SEEN.` and click `testid=submit-button`. The user bubble contains `image-carousel` / `image-preview`. Run `control-openhands browser hover 'testid=image-preview'`, `control-openhands browser click 'testid=expand-image-button'` (`count 'testid=image-lightbox'` is `1`), `control-openhands browser press Escape` (`0`), click `expand-image-button` again and `control-openhands browser click 'testid=image-lightbox-close'` (`0`).@@ -101,6 +101,42 @@ Preconditions:- **Phone (`F06.phone`).** Run `control-openhands browser viewport phone`, open the table conversation (`control-openhands browser goto /conversations/<stream id>`), `control-openhands browser bbox 'testid=chat-interface'` (`insideViewport` `true`, `pageHorizontalOverflow` `false`) and `control-openhands browser bbox 'testid=markdown-table-scroll'` (`scrollWidth` larger than `clientWidth`: the table scrolls inside itself). `control-openhands browser screenshot --feature F06.phone --name conversation`, then `control-openhands browser viewport desktop`.- After each group run `control-openhands browser errors --app-only`. Outages driven with `service stop` log connection-refused console errors by design; `pageErrors` must stay `0`.+- **Planner discovery with many children (`F06.tool-visualizers`).** Expected: reloading a parent still discovers its `plannerparent`-tagged planner when it has100 child conversations. Known failure (reproduced 2026-10-16, SDK 1.55.0, [OpenHands/software-agent-sdk#5790](https://github.com/OpenHands/software-agent-sdk/issues/5790)): the unchunked100-ID request returns500, shows `Internal Server Error` and omits the planner session connection. A99-child control returns200 and opens both planner and parent sockets. These are genuine idle API-created children, without a model run or browser-storage changes; a new parent ID has no UI-created planner hint.+ ```sh known-bug F06.tool-visualizers planner-batch-100+ control-openhands browser viewport desktop+ QA_ID="$(python3 - <<'PY'+ import json, os, pathlib, subprocess, tempfile+ base = pathlib.Path(os.environ["OH_VERIFY_RUN"]) / "workspace"+ base.mkdir(exist_ok=True)+ fixture = pathlib.Path(tempfile.mkdtemp(prefix="qa-surface-", dir=base))+ llm = {"model":"openai/gpt-4o", "base_url":"http://127.0.0.1:1/v1", "api_key":"qa-not-real", "num_retries":0, "timeout":2}+ def create(workspace, parent=None, tags=None):+ payload = {"agent":{"kind":"Agent", "llm":llm, "tools":[]}, "workspace":{"working_dir":str(workspace)}, "autotitle":False}+ if parent: payload["parent_conversation_id"] = parent+ if tags: payload["tags"] = tags+ request = fixture / "request.json"+ request.write_text(json.dumps(payload))+ result = subprocess.run(["control-openhands", "api", "POST", "/api/conversations", "--write", "--data", "@"+str(request), "--pick", "id"], check=True, capture_output=True, text=True)+ return json.loads(result.stdout)["value"]+ def git(directory, *args):+ subprocess.run(["git", "-C", str(directory), *args], check=True, capture_output=True)+ workspace = fixture / "workspace"+ workspace.mkdir()+ parent = create(workspace)+ for index in range(100):+ create(workspace, parent, {"plannerparent":parent} if index == 0 else None)+ print(parent)+ PY+ )"+ control-openhands assert --field value --equals 100 -- api GET "/api/conversations/$QA_ID" --pick sub_conversation_ids.length+ QA_PLANNER="$(control-openhands api GET "/api/conversations/$QA_ID" --pick sub_conversation_ids.0 | python3 -c 'import json,sys; print(json.load(sys.stdin)["value"])')"+ control-openhands browser network --clear+ control-openhands browser goto "/conversations/$QA_ID"+ control-openhands browser wait 'testid=chat-input'+ control-openhands browser screenshot --feature F06.tool-visualizers --name planner-batch-100+ control-openhands assert --field recent.0.status --equals 200 --bug-shows 500 -- browser network --filter "ids=$QA_PLANNER" # bug+ ```+## Gotchas- `control-openhands conversation events <id>` returns the last 25 events unless you pass `--last N`; it reads the server's 100-event pages until `N` rows are in hand (`pages` says how many it read, `more: true` that older pages remain), so `--last 500` covers a 150-event conversation in two pages (`total` `152`, `pages` `2`, `more` `false` on the Older history conversation; `--last 25` there reads one page, `more` `true`) and `--from-start` reads from the oldest. `--kinds` and `--grep` filter before `--last` counts, and pages are read until `--last` matching rows are in hand: `--kinds MessageEvent --last 3` is the three newest messages, `--kinds MessageEvent --last 500` all thirty, and `--grep 'qa older 1'` reaches `qa older 1` on the oldest page with the default `--last` (`count` `11`, `pages` `2`). Event texts are cut at about 160 characters.@@ -125,3 +161,52 @@ Preconditions:- The no-LLM banner reuses the home banner's test ids (`home-llm-not-configured-banner`). It exists only in a run where no LLM was ever configured: use a separate `launch --new`, never remove the profile from the shared run.- Read the clipboard with `control-openhands browser clipboard`, never `browser eval "navigator.clipboard.readText()"`. An agent bubble that contains a code block holds two `copy-to-clipboard` buttons (the code block's is hidden until hovered): scope the bubble's with `>> nth=0`.- Plan-mode previews (`plan-preview-*`) and the BTW side messages render in this stream but are reached from the composer's mode controls; they are not mapped here.+- Known issue [#18117](https://github.com/OpenHands/OpenHands/issues/18117) (`F06.thinking`): Inline thinking leaks into the agent chat bubble. Live prerequisite: A live model that emits a leading inline think block on a tool-call turn, plus its finalized ActionEvent.+- Known issue [#17559](https://github.com/OpenHands/OpenHands/issues/17559) (`F06.load-older-history`): History pagination can show both an action and its observation (ready-for-dev companion to #16947). Live prerequisite: A real persisted history page boundary splitting an action/observation or ACP started/completed pair; dummy-key errors do not produce these tool pairs.+- Known issue [#17492](https://github.com/OpenHands/OpenHands/issues/17492) (`F06.tool-visualizers`): Failed browser, glob, grep, and file editor tool calls render as successful in chat. Live prerequisite: A working model emitting failed browser, glob, grep, planning-editor and file-editor observations with is_error=true.+- Known issue [#17490](https://github.com/OpenHands/OpenHands/issues/17490) (`F06.tool-visualizers`): Terminal commands that fail with an ordinary non-zero exit render as successful in chat. Live prerequisite: A working model producing a completed terminal command with nonzero exit_code and is_error=false.+- Known issue [#17161](https://github.com/OpenHands/OpenHands/issues/17161) (`F06.markdown-rendering`): markdown sanitizer lets untrusted content set rel="opener" and any target on links. Live prerequisite: A real reasoning, activated-skill, tool-description or error-detail body rendered without includeStandard. The attempted skills dialog exposed public skills only and did not reach the affected chat renderer.+- Known issue [#17026](https://github.com/OpenHands/OpenHands/issues/17026) (`F06.thinking`): agent-canvas: final-reply thinking block missing for text-only turns (reasoning_content not read for MessageEvents). Live prerequisite: A reasoning-capable live model emitting nonempty MessageEvent.llm_message.reasoning_content on a text-only turn, followed by reload.+- Known issue [#16842](https://github.com/OpenHands/OpenHands/issues/16842) (`F06.reconnect`): A late close from a replaced WebSocket marks the active connection disconnected. Live prerequisite: A transport fixture delaying an old socket close until its replacement is open; ordinary server restart cannot select that ordering.+- Known issue [#16719](https://github.com/OpenHands/OpenHands/issues/16719) (`F06.live-activity`): Agent status always says "thinking" even when nothing is happening. Live prerequisite: The event/status trace and deterministic stalled or errored state from the original report; its trigger is unspecified.+- Known issue [#16686](https://github.com/OpenHands/OpenHands/issues/16686) (`F06.error-events`): LLM errors not propagated to the frontend. Live prerequisite: A Nebius credential and its rejected DeepSeek-V3 request, or the original complete emitted error fixture.+- Known issue [#16582](https://github.com/OpenHands/OpenHands/issues/16582) (`F06.stop-resume`): After reconnect, a maximum-iteration error can fail to pause the agent through the legacy event handler. Live prerequisite: A maximum-iteration legacy error delivered after socket replacement, with a real model run reaching that limit.+- Known issue [#16455](https://github.com/OpenHands/OpenHands/issues/16455) (`F06.events-match`): Duplicate messages from the agent in 1.12.0. Live prerequisite: The original duplicated event/stream trace or deterministic prompt and timing; the report provides no reproducible trigger.+- Known issue [#16341](https://github.com/OpenHands/OpenHands/issues/16341) (`F06.tool-visualizers`): Code-explorer subagent frequently errors. Live prerequisite: A working model with native code-explorer delegation and the original repository task; no delegated-tool output can be generated without credentials.+- Known issue [#15626](https://github.com/OpenHands/OpenHands/issues/15626) (`F06.error-events`): Conversation can fail with LiteLLM auth error when database server at 127.0.0.1:5432 is unreachable. Live prerequisite: The original LiteLLM proxy/database-failure event or a disposable proxy with its PostgreSQL dependency unavailable.+- Known issue [#15606](https://github.com/OpenHands/OpenHands/issues/15606) (`F06.reload-mid-run`): When re-loading ACP conversations, messages disappear or move. Live prerequisite: An authenticated Codex ACP session emitting intermediate messages while another window reloads history.+- Known issue [#15527](https://github.com/OpenHands/OpenHands/issues/15527) (`F06.reconnect`): Canvas briefly shows "disconnected" when coming back to an idle tab. Live prerequisite: The original custom-domain deployment, idle duration and tab-suspension/network transition; those details are missing.+- Known failure [OpenHands/software-agent-sdk#5635](https://github.com/OpenHands/software-agent-sdk/issues/5635) (`F06.reconnect`), reproduced 2026-10-12 at `fa1f724ab573` with SDK 1.54.0 and Chromium 153.0.8010.12: sandbox pause returned 204 but an already communicating session socket remained open for more than 10 seconds. A new Canvas tab received a subsequent `run:false` marker while the original tab did not. DELETE returned 200 and GET returned 404, yet both held sockets remained open for more than 10 seconds; Canvas later redirected by metadata polling. Recipe: open an idle dummy-profile conversation and confirm its marker/socket; POST `/api/conversations/prepare-for-sandbox-pause`, retain the original tab for 10 seconds, open a second tab, POST a new unique user event with `run:false`, then compare the two transcripts. DELETE the owned conversation and read lifecycle metadata for another 10 seconds. Use `--run <run>` and `browser events --kinds websocket-created,websocket-open,websocket-close --last 100`; no frame bodies or auth data are recorded. Evidence: `backfill-18309-surface-rechecks/F06.reconnect/pause-delete-sockets-5635.json`, `pause-new-tab-5635.png`, `pause-stale-tab-5635.png`. The expected socket shutdown/reconnect remains unchanged; polling recovery must not be mistaken for socket closure.+- Known issue [OpenHands/software-agent-sdk#5465](https://github.com/OpenHands/software-agent-sdk/issues/5465) (`F06.branch-from-here`): Branch from here can lose encrypted LLM credentials, so the fork fails its first turn with LLMAuthenticationError. Live prerequisite: A working cipher-backed model conversation, then fork and a successful turn on the fork.+- Known issue [OpenHands/software-agent-sdk#5447](https://github.com/OpenHands/software-agent-sdk/issues/5447) (`F06.error-events`): An Anthropic Claude Opus5.5 conversation can fail with400 when its next request ends in an assistant prefill. Live prerequisite: Anthropic Claude Opus5.5 credentials and a real trailing-assistant-prefill request.+- Known issue [OpenHands/software-agent-sdk#5354](https://github.com/OpenHands/software-agent-sdk/issues/5354) (`F06.event-groups`): A mixed valid/invalid tool batch can be split into separate assistant messages, breaking call/result pairing on the next provider request. Live prerequisite: A live model turn mixing valid, invalid and valid tool calls in one response.+- Known issue [OpenHands/software-agent-sdk#5225](https://github.com/OpenHands/software-agent-sdk/issues/5225) (`F06.error-events`): A provider-rejected image tool result can leave the conversation permanently failing on every resume. Live prerequisite: A real provider-rejected poisoned image tool_result in a disposable persisted history, followed by retry.+- Known issue [OpenHands/software-agent-sdk#5108](https://github.com/OpenHands/software-agent-sdk/issues/5108) (`F06.tool-visualizers`): The native SDK Gemini editor can report success while corrupting unrelated invalid-UTF8 bytes in the file. Live prerequisite: An Agent Server configured with the native Gemini edit tool set and a real tool edit of an invalid-UTF8 fixture; Gemini ACP uses a different external executor.+- Known issue [OpenHands/software-agent-sdk#5096](https://github.com/OpenHands/software-agent-sdk/issues/5096) (`F06.tool-visualizers`): The canonical agent file editor can replace a symlink instead of editing its target; Undo does not restore the link. Live prerequisite: A working model invoking canonical file-editor str_replace/insert through a real symlink, then undo.+- Known issue [OpenHands/software-agent-sdk#5092](https://github.com/OpenHands/software-agent-sdk/issues/5092) (`F06.confirmation-mode`): A revocation message sent while waiting for confirmation can implicitly approve the pending action. Live prerequisite: A working model proposing a real confirmation-gated marker action, then a revocation user message.+- Known issue [OpenHands/software-agent-sdk#5074](https://github.com/OpenHands/software-agent-sdk/issues/5074) (`F06.tool-visualizers`): After server restart a native delegated child can remain Running while the parent loses its task mapping/result. Live prerequisite: A native delegated child still running during an owned server crash and parent recovery.+- Known issue [OpenHands/software-agent-sdk#5038](https://github.com/OpenHands/software-agent-sdk/issues/5038) (`F06.tool-visualizers`): The agent file editor can refuse valid Chinese UTF8 text as binary when its sample splits a character; Canvas Files uses a separate preview detector. Live prerequisite: A live canonical file-editor view/edit on the specific UTF8 sample alignment; Canvas Files preview uses a separate NUL sniff, not this validator.+- Known issue [OpenHands/software-agent-sdk#4990](https://github.com/OpenHands/software-agent-sdk/issues/4990) (`F06.stop-resume`): An ACP turn can keep executing tool calls while the UI already reports Paused. Live prerequisite: An authenticated Claude Code ACP turn executing separate marker-writing tools while Pause is pressed.+- Known issue [OpenHands/software-agent-sdk#4983](https://github.com/OpenHands/software-agent-sdk/issues/4983) (`F06.tool-visualizers`): Agent grep can silently miss extended-regex matches when its runtime falls back to system grep. Live prerequisite: A runtime without ripgrep and a live agent grep call using extended regex and escaped literal punctuation.+- Known issue [OpenHands/software-agent-sdk#4965](https://github.com/OpenHands/software-agent-sdk/issues/4965) (`F06.error-events`): A strict OpenAI-compatible provider can reject blank content in the saved history, preventing later turns. Live prerequisite: A strict OpenAI-compatible provider and a real history containing blank text or empty content arrays.+- Known issue [OpenHands/software-agent-sdk#4777](https://github.com/OpenHands/software-agent-sdk/issues/4777) (`F06.stop-resume`): Stop can still start a queued tool after it finishes waiting on a resource lock. Live prerequisite: A live batch with a second real tool waiting on a declared-resource lock when Stop is pressed.+- Known issue [OpenHands/software-agent-sdk#4668](https://github.com/OpenHands/software-agent-sdk/issues/4668) (`F06.tool-visualizers`): A planning editor card can lose old/new content and fail to show the complete edit diff. Live prerequisite: A live planner create/edit yielding PlanningFileEditorObservation with old/new contents for the chat diff.+- Known issue [OpenHands/software-agent-sdk#4663](https://github.com/OpenHands/software-agent-sdk/issues/4663) (`F06.tool-visualizers`): Parallel Python-fallback glob calls can change process cwd and return files from the wrong workspace. Live prerequisite: Concurrent real glob/tool calls in distinct workspaces with the Python fallback active.+- Known issue [OpenHands/software-agent-sdk#4661](https://github.com/OpenHands/software-agent-sdk/issues/4661) (`F06.tool-visualizers`): Same-name project subagents can resolve to another conversation workspace definition. Live prerequisite: Two native task-enabled conversations with same-name project subagent definitions and a working model for delegation.+- Known issue [OpenHands/software-agent-sdk#4583](https://github.com/OpenHands/software-agent-sdk/issues/4583) (`F06.tool-visualizers`): An agent editor insertion at EOF can join onto the last line when the file has no trailing newline. Live prerequisite: A working model invoking file-editor insert at EOF on a file with no trailing newline.+- Known issue [OpenHands/software-agent-sdk#4558](https://github.com/OpenHands/software-agent-sdk/issues/4558) (`F06.error-events`): A saved encrypted fallback profile can send ciphertext as its key and fail authentication, leaving only the primary error visible. Live prerequisite: A working encrypted saved fallback profile and real primary-model failure that triggers fallback resolution.+- Known issue [OpenHands/software-agent-sdk#4540](https://github.com/OpenHands/software-agent-sdk/issues/4540) (`F06.tool-visualizers`): Synthetic GLM/Kimi tool calls can render as raw tool_call syntax instead of executing. Live prerequisite: Synthetic provider credentials and a GLM/Kimi turn emitting the raw tool_call wrapper.+- Known issue [OpenHands/software-agent-sdk#4532](https://github.com/OpenHands/software-agent-sdk/issues/4532) (`F06.confirmation-mode`): After restarting while awaiting confirmation, approval can execute the tool but the next model request omits its assistant call. Live prerequisite: A real confirmation-paused action persisted across backend restart, then approval and a working provider call.+- Known issue [OpenHands/software-agent-sdk#4077](https://github.com/OpenHands/software-agent-sdk/issues/4077) (`F06.reload-mid-run`): Retried or late streaming deltas can duplicate live text; a stalled subscriber can delay Stop and finalization. Live prerequisite: A working stream interrupted/retried or delivered to a controlled stalled subscriber, with delta/state-order capture.+- Known issue [#16947](https://github.com/OpenHands/OpenHands/issues/16947) (`F06.load-older-history`): Loading older history can leave both the action and its newer observation visible because UI projection depends on arrival order. Live prerequisite: A real long conversation whose REST boundary splits an action/observation pair; keep companion #17559 linked separately.++- Known issue [#18445](https://github.com/OpenHands/OpenHands/issues/18445) (`F06.load-older-history`, `F06.reconnect`): a failed initial Cloud history search can be swallowed into an empty page while the runtime count succeeds, advancing `after_seq` past history that was never displayed. Live prerequisite: a Cloud account and owned conversation with a genuinely failing App API `events/search` and a successful runtime `events/count`. The Local failure path throws instead and is not an equivalent check.++- Known issue [OpenHands/software-agent-sdk#5969](https://github.com/OpenHands/software-agent-sdk/issues/5969) (`F06.stop-resume`): a conversation can report finished but reject Play with already-running, or delay a follow-up, while a slow event webhook is still pending. Live prerequisite: a genuinely successful model run on an owned backend with a bounded slow webhook receiver, then Play or send during that completion window.++- Known issue [OpenHands/software-agent-sdk#5740](https://github.com/OpenHands/software-agent-sdk/issues/5740) (`F06.error-events`): a Responses display delta without `item_id` can terminate the turn with an AttributeError. Live prerequisite: the genuine provider/transformer streaming path producing that incomplete event; the report did not retain its original raw SSE. A credential-free parser diagnostic proves the SDK exception, not a Canvas model turn.++- Known issue [OpenHands/software-agent-sdk#5806](https://github.com/OpenHands/software-agent-sdk/issues/5806) (`F06.branch-from-here`): editing a displayed user message first reads its parent event; if that event has become unreadable/missing on the server, the action can fail with 500 instead of a missing-event response. Live prerequisite: an owned persisted conversation with a still-displayed user message whose event storage is missing, followed by the actual Edit action. A random unknown-ID API call proves neither this UI path nor the batch-get contract, which Canvas does not call.++- Session reconnects use the latest received `after_seq` cursor (`F06.reconnect`). For a local metadata check, retain the open page, send genuine `run:false` user events, record `/events/count`, restart the owned backend and inspect finite WebSocket lifecycle metadata: the reconnect cursor is the previous count minus one and a subsequent recovery marker appears once. This does not measure frame-body replay or Cloud behavior.+- Stop persists `paused` before an in-flight model request completes (`F06.stop-resume`); the persisted state and reloaded Stopped UI survive an Agent Server restart. A bounded credential-free check can use the tracked accept-only endpoint and an exact listener PID proven to descend from the owned launcher; never kill by pattern or resume the paused conversation.diff --git a/.agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md b/.agents/skills/verify-openhands/references/feature-map/F07-conversation-page.mdindex 992bf86aa5..eb1f17850c 100644--- a/.agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md+++ b/.agents/skills/verify-openhands/references/feature-map/F07-conversation-page.md@@ -86,9 +86,45 @@ Preconditions:- **Cloud-only (`F07.cloud-only`).** Blocked without a Cloud backend. On local, confirm the absence: the menu testids (see `F07.menu-open`) have no `share-publicly-button`, and the git control bar shows no `Connect Repo` chip. With Cloud: the menu's **Share publicly** switch (`share-publicly-button`, then `copy-share-link-button`, `open-share-link-button`), the repo chip **Connect Repo** opening the Open repository modal (`close-open-repository-modal`), `/conversations/task-<uuid>` while a conversation starts, and the `Unavailable for Archives` tooltips (see F25).- **Phone (`F07.phone`).** Run `control-openhands browser viewport phone`, `control-openhands browser testids 'testid=chat-pane-header'` (adds `sidebar-mobile-menu-toggle`; all header controls visible), open each info dialog from the menu and `control-openhands browser bbox 'testid=skills-modal'` (likewise `hooks-modal`, `system-message-modal`): `insideViewport` is `true` and `pageHorizontalOverflow` `false`. Then `control-openhands browser click 'testid=chat-pane-header >> testid=ellipsis-button'` and `control-openhands browser bbox 'testid=conversation-name-context-menu'`: inside the viewport too (`x` about 145, `width` about 237), and `control-openhands browser screenshot --feature F07.phone --name header-menu` shows every label whole (`Show Available Skills`, `Download conversation data`, `Delete Conversation`). `control-openhands browser press Escape` closes it. Return with `control-openhands browser viewport desktop`.+- **Whitespace-only header rename (`F07.rename-inline`).** After the named-title check, run `control-openhands browser network --clear`, double-click `testid=conversation-name-title`, fill `testid=conversation-name-input` with one space, and `control-openhands browser press Enter`. Read `control-openhands browser network --filter '/api/conversations' --bodies`, then reload and read the title. Expected: restore the previous title with no write. Known failure (reproduced 2026-10-12, [#17061](https://github.com/OpenHands/OpenHands/issues/17061)): Canvas submits `PATCH` with an empty title. Agent Server 1.54.0 rejects it with 422 and restores the prior title; a persistently blank server title was not observed.++- **Already-deleted conversation (`F07.delete-confirm`, `F07.rename-inline`).** Expected: a stale action reports that the conversation is missing, and a stale rename never reports success. Known failure (reproduced 2026-10-16, SDK 1.55.0, [OpenHands/software-agent-sdk#5800](https://github.com/OpenHands/software-agent-sdk/issues/5800)): confirming Delete after another client removed this disposable conversation returns400 and displays `Bad Request`, instead of a missing-conversation404. The separate stale-rename control receives PATCH200 but re-fetches missing metadata and displays not-found, so its false-success variant is prevented by Canvas. Pause/interrupt were not tested.+ ```sh known-bug F07.delete-confirm already-deleted+ control-openhands browser viewport desktop+ QA_ID="$(python3 - <<'PY'+ import json, os, pathlib, subprocess, tempfile+ base = pathlib.Path(os.environ["OH_VERIFY_RUN"]) / "workspace"+ base.mkdir(exist_ok=True)+ fixture = pathlib.Path(tempfile.mkdtemp(prefix="qa-surface-", dir=base))+ llm = {"model":"openai/gpt-4o", "base_url":"http://127.0.0.1:1/v1", "api_key":"qa-not-real", "num_retries":0, "timeout":2}+ def create(workspace, parent=None, tags=None):+ payload = {"agent":{"kind":"Agent", "llm":llm, "tools":[]}, "workspace":{"working_dir":str(workspace)}, "autotitle":False}+ if parent: payload["parent_conversation_id"] = parent+ if tags: payload["tags"] = tags+ request = fixture / "request.json"+ request.write_text(json.dumps(payload))+ result = subprocess.run(["control-openhands", "api", "POST", "/api/conversations", "--write", "--data", "@"+str(request), "--pick", "id"], check=True, capture_output=True, text=True)+ return json.loads(result.stdout)["value"]+ def git(directory, *args):+ subprocess.run(["git", "-C", str(directory), *args], check=True, capture_output=True)+ workspace = fixture / "workspace"+ workspace.mkdir()+ print(create(workspace))+ PY+ )"+ control-openhands browser goto "/conversations/$QA_ID"+ control-openhands browser click 'testid=chat-pane-header >> testid=ellipsis-button'+ control-openhands browser click 'testid=conversation-name-context-menu >> testid=delete-button'+ control-openhands assert --field status --equals 200 -- api DELETE "/api/conversations/$QA_ID" --write+ control-openhands browser network --clear+ control-openhands browser click 'role=dialog >> testid=confirm-button' --observe 'role=status' --observe-ms 1000+ control-openhands browser screenshot --feature F07.delete-confirm --name stale-delete+ control-openhands assert --field recent.0.status --equals 404 --bug-shows 400 -- browser network --filter "/api/conversations/$QA_ID" # bug+ ```+## Gotchas-- The stop and delete confirmation buttons carry `cancel-button` and `confirm-button` (#17915, fixed by #18151); role names inside `role=dialog` work too.+- The stop and delete confirmation buttons carry `cancel-button` and `confirm-button`; role names inside `role=dialog` work too.- `stop-button`, `show-skills-button`, `show-hooks-button` and `show-agent-tools-button` also exist in the composer, and `ellipsis-button` in sidebar rows: always scope to `testid=chat-pane-header` or `testid=conversation-name-context-menu`.- Stop Runtime (status menu) and Stop Conversation (menu) both navigate to `/conversations`. On a finished local conversation the agent server keeps `finished`; use a running conversation to see `paused` and Start Runtime.- Rename has no Escape-to-cancel: Escape leaves the input open and the next blur saves whatever was typed.@@ -109,6 +145,20 @@ Preconditions:- Dialogs move focus inside when they open and keep Tab inside: the Skills and Hooks dialogs start on their close button, Stop and Delete on `cancel-button`, Export on `close-transcript-export-modal`, Agent Tools on `close-system-message-modal`. In the Hooks dialog, focusing Refresh with Tab opens its tooltip, which takes the first Escape; a second Escape closes the dialog.- The Hooks dialog shows the workspace's `hooks.json` as of its last read (its first open for the conversation after a page load, or Refresh), not the hooks the conversation started with: after the hooks steps above, `browser goto /conversations/<git-id>` (started before the file existed; its `hook_config` is `null`) and opening the dialog lists both events too. A reopen within 5 minutes of that read shows the same list and sends no request: click Refresh after editing `hooks.json`.- A prompt- or agent-type hook shows its text under Commands (the `prompt`, or an agent hook's `system_prompt`). With the Hooks dialog closed (the hooks steps above end that way; otherwise `control-openhands browser click 'testid=close-hooks-modal'`), write `<run>/workspace/qa-f07-repo/.openhands/hooks.json` as `{"stop":[{"hooks":[{"type":"prompt","prompt":"QA_F07 prompt hook"}]}],"session_start":[{"hooks":[{"type":"agent","system_prompt":"QA_F07 agent hook"}]}]}`, run `control-openhands browser goto /conversations/<hooks-id>`, open the menu and `control-openhands browser click 'testid=conversation-name-context-menu >> testid=show-hooks-button'`, click `testid=hooks-modal >> testid=refresh-hooks`, then `role=button[name="Session Start 1 hook"]` and `role=button[name="Stop 1 hook"]` inside `testid=hooks-modal` (an event expanded before a Refresh stays expanded, and a second click collapses it, so start from a reopened dialog). In `control-openhands browser snapshot 'testid=hooks-modal'` they expand, in that order, to `Matcher * Commands QA_F07 agent hook Type: agent Timeout: 60s` and `Matcher * Commands QA_F07 prompt hook Type: prompt Timeout: 60s`, and `browser text 'testid=hooks-modal >> pre'` returns `QA_F07 agent hook` and `QA_F07 prompt hook` (`control-openhands browser screenshot --feature F07.hooks-modal --name prompt-agent-hooks`). An agent hook without `system_prompt` shows no box: with `{"session_start":[{"hooks":[{"type":"agent"}]}]}`, close, reopen and Refresh, `Session Start 1 hook` expands to `Matcher * Commands Type: agent Timeout: 60s` and `browser count 'testid=hooks-modal >> pre'` is `0`.-- An agent hook shows its `system_prompt`, the text it runs with, even when it also carries a `prompt`, and an agent hook with only a `prompt` shows no box (#18206, fixed by #18215). With `{"session_start":[{"hooks":[{"type":"agent","prompt":"QA_F07 stray prompt","system_prompt":"QA_F07 agent system prompt"}]}],"stop":[{"hooks":[{"type":"agent","prompt":"QA_F07 prompt only"}]}]}` in `hooks.json`, close, reopen and Refresh the dialog, then expand both events: `Session Start 1 hook` reads `Matcher * Commands QA_F07 agent system prompt Type: agent Timeout: 60s`, `Stop 1 hook` reads `Matcher * Commands Type: agent Timeout: 60s`, and `control-openhands browser count 'text=QA_F07 stray prompt'` and `browser count 'text=QA_F07 prompt only'` are `0` (`browser screenshot --feature F07.hooks-modal --name agent-hook-system-prompt`). Agent hooks in `hooks.json` really run on the next turn in that workspace (the session-start one shows a `Hook: SessionStart failed` row), so write the recipe's two-event `hooks.json` (`pre_tool_use` and `stop`) back and Refresh afterwards.+- An agent hook shows its `system_prompt`, the text it runs with, even when it also carries a `prompt`, and an agent hook with only a `prompt` shows no box. With `{"session_start":[{"hooks":[{"type":"agent","prompt":"QA_F07 stray prompt","system_prompt":"QA_F07 agent system prompt"}]}],"stop":[{"hooks":[{"type":"agent","prompt":"QA_F07 prompt only"}]}]}` in `hooks.json`, close, reopen and Refresh the dialog, then expand both events: `Session Start 1 hook` reads `Matcher * Commands QA_F07 agent system prompt Type: agent Timeout: 60s`, `Stop 1 hook` reads `Matcher * Commands Type: agent Timeout: 60s`, and `control-openhands browser count 'text=QA_F07 stray prompt'` and `browser count 'text=QA_F07 prompt only'` are `0` (`browser screenshot --feature F07.hooks-modal --name agent-hook-system-prompt`). Agent hooks in `hooks.json` really run on the next turn in that workspace (the session-start one shows a `Hook: SessionStart failed` row), so write the recipe's two-event `hooks.json` (`pre_tool_use` and `stop`) back and Refresh afterwards.- With the Agent Server down, Refresh in the Hooks dialog shows its own fetch error (#18177): after `control-openhands service stop agent-server` on an open dialog, `refresh-hooks` and `browser wait 'testid=hooks-modal >> text=Failed to fetch hooks' --timeout 30000`, the dialog reads `Failed to fetch hooks. Please try again later.` under its explanation (`browser count 'testid=hooks-modal >> text=Failed to fetch skills'` is `0`). The 502s and the closed WebSocket that `browser errors --app-only` then lists are the outage itself. Run it last, or `restart` afterwards.- After the user-message branch in `F07.branch-from-message`, Send stays disabled next to the prefilled text (`browser enabled 'testid=submit-button'` is `false`; Known failure #18202, see F05 Gotchas). Enter in the field sends it.+- Known issue [#18136](https://github.com/OpenHands/OpenHands/issues/18136) (`F07.cloud-only`): Connect Repo on a cloud conversation does not save the repository on the server. Live prerequisite: An authenticated Cloud sandbox/repository provider and a second fresh browser to compare server repository metadata.+- Known issue [#17494](https://github.com/OpenHands/OpenHands/issues/17494) (`F07.export-transcript`): Transcript export leaks raw i18n template for ACP tool calls without a title. Live prerequisite: An authenticated ACP session that emits a titleless tool call, then Markdown and HTML export.+- Known issue [#16876](https://github.com/OpenHands/OpenHands/issues/16876) (`F07.history-skeleton`): Planning-agent history flag stays true forever when sub-conversations fetch fails — infinite chat loading skeleton. Live prerequisite: A real planning sub-conversation whose batch lookup fails or yields null, with the main history available.+- Known issue [#16727](https://github.com/OpenHands/OpenHands/issues/16727) (`F07.open-by-url`): Conversation corrupted after VM restart: "Missing event file". Live prerequisite: A disposable VM interrupted at the event-write boundary or the original missing-event persisted fixture.+- Known issue [#16356](https://github.com/OpenHands/OpenHands/issues/16356) (`F07.open-by-url`): Conversation LLM config is frozen at creation time — resuming after API key rotation fails permanently with LLMAuthenticationError. Live prerequisite: A working provider key that can be rotated/revoked, an existing conversation using it, and a replacement profile.+- Known issue [#15543](https://github.com/OpenHands/OpenHands/issues/15543) (`F07.git-control-bar`): Bad parsing for branch retrieval. Live prerequisite: The original Git probe stdout/interleaving that temporarily puts a filename in the branch badge; dirty state alone is not a deterministic trigger.+- Known issue [OpenHands/software-agent-sdk#5640](https://github.com/OpenHands/software-agent-sdk/issues/5640) (`F07.download-zip`): Download conversation data can return404 after a warm-pool server initializes a custom conversation-storage path. Although closed upstream, the pinned SDK 1.55.0 still reads the default configuration path here; this remains a supported-version distinction, not a freshly reproduced failure. Live prerequisite: A deferred-init warm-pool Agent Server initialized with a non-default conversations_path; this launcher starts an already-initialized server.+- Known failure [OpenHands/software-agent-sdk#5629](https://github.com/OpenHands/software-agent-sdk/issues/5629) (`F07.rename-inline`), reproduced 2026-10-12 at `fa1f724ab573`, SDK 1.54.0. A second server's native first load rewrote a newer saved title with its startup copy; the connected Canvas header showed that older title. Recipe: create an idle conversation on A and PATCH its title to `QA_SHARED_BEFORE_B_5629_REPEAT`; run pinned SDK Python `python .agents/skills/verify-openhands/scripts/fixtures/shared-storage-server.py --run <run> --seconds 300` in the background. It starts only a bounded no-editor SDK B sharing this run's conversation directory/cipher/session key; its private `server.json` supplies the port and owned operator/PGID. After B listens, PATCH A to `QA_SHARED_AFTER_B_5629_REPEAT` and confirm disk. Run `control-openhands service stop agent-server --run <run>`, then `python .../fixtures/shared-storage-server.py --run <run> --load-id <id>`: native B `/events/count` returned 200 and disk reverted to BEFORE_B. Restart A only to serve Canvas, choose default-local and doctor; use the Add Backend recipe from [F25](F25-backends-and-cloud.md) to connect B at its recorded port with the private key file, then open its card: the header reads BEFORE_B. SIGTERM only the recorded helper operator and verify its port/PGID gone; the helper also bounds runtime and terminates remaining group descendants. Evidence: `backfill-18309-surface-rechecks/F07.rename-inline/metadata-repeat-before-b-load-5629.json`, `metadata-repeat-after-b-load-5629.json`, `shared-first-load-reverted-5629.png`. The native first-load request is explicit; stale catalog display alone is insufficient proof of this overwrite.+- Known failure [OpenHands/software-agent-sdk#5539](https://github.com/OpenHands/software-agent-sdk/issues/5539) (`F07.download-zip`), reproduced 2026-10-12 at `fa1f724ab573`, SDK 1.54.0: three real overlapping receivers of one idle conversation made 929 requests in 20 seconds, including 31 completed corrupt ZIPs with status 200, 106 server 500s, 269 ingress 502s and 133 RemoteProtocolErrors. A sequential 2,000-request control returned valid ZIPs. Recipe (use the pinned SDK Python with httpx): `python .agents/skills/verify-openhands/scripts/fixtures/download-trajectory-overlap.py --base-url <own-base-url> --conversation-id <id> --api-key-file <run>/private/session-key --output-dir <run>/private/trajectory-overlap --seconds 20 --clients 3 --rounds 5000 --slow-ms 2`. It enforces loopback, request/duration/network/uncompressed/member budgets and bounded CRC reads; aborts and unsupported archives remain unverified. While it runs, use `browser goto /conversations/<id>`, open `chat-pane-header >> ellipsis-button`, click `download-trajectory-button`, then `browser downloads --last 3 --inspect` (all CLI commands with `--run <run>`). That Canvas click downloaded an inspectable ZIP; its particular result is not claimed corrupt. Private raw samples stay private. Evidence: `backfill-18309-surface-rechecks/F07.download-zip/trajectory-overlap-5539-summary.json` and `overlapping-download-5539.png`. The summary correction excludes two ambiguous Content-Length differences that could be intentional deadline aborts; only completed corrupt ZIPs count as corrupt downloads.+- Known issue [OpenHands/software-agent-sdk#5094](https://github.com/OpenHands/software-agent-sdk/issues/5094) (`F07.open-by-url`): After ACP session/load fails, the backend can create a fresh model session while Canvas still shows the old context. Live prerequisite: An authenticated Codex ACP session whose original session/load fails while persisted history remains.+- Known issue [OpenHands/software-agent-sdk#4893](https://github.com/OpenHands/software-agent-sdk/issues/4893) (`F07.status-menu`): A sandbox frozen mid-run can resume with a stale Running state that rejects normal resume requests. Live prerequisite: A disposable sandbox/VM frozen mid-run with a live model and persisted ownership state.+- Known failure [OpenHands/software-agent-sdk#4800](https://github.com/OpenHands/software-agent-sdk/issues/4800) (`F07.rename-inline`), reproduced 2026-10-12 at `fa1f724ab573`, SDK 1.54.0. A refused rename changed the server's GET title while `meta.json` kept the previous title; graceful shutdown subsequently persisted the rejected title. Recipe: create an idle dummy-profile conversation, then double-click `conversation-name-title`, fill `conversation-name-input` with `QA_META_BEFORE_4800` and press Enter; verify GET/disk. Record the original mode and temporarily `chmod 0500 <run>/private/state/dev_conversations/<id-hex>` to deny an atomic metadata replacement. Rename through that UI to `QA_META_REFUSED_4800`: the permission-denied toast/HTTP 500 appears, GET shows REFUSED, disk still shows BEFORE. Restore the exact original directory mode in `finally`, then `restart`, `doctor`, GET and reload (CLI with `--run <run>`): REFUSED was persisted during graceful shutdown. Only the owned temporary directory is changed; evidence contains title/timestamp fields, not stored secrets. Evidence: `backfill-18309-surface-rechecks/F07.rename-inline/metadata-before-4800.json`, `metadata-denied-4800.json`, `refused-title-4800.png`, `metadata-after-graceful-restart-4800.json`. Tags/autotitle and ownership-conflict branches remain untested.++- Known issue [#16877](https://github.com/OpenHands/OpenHands/issues/16877) (`F07.open-by-url`): Untitled terminal-status conversations are reported to keep the 3-second title polling interval. On frozen target `fa1f724ab5737a17453ad6081129eb91c5a423c6`, a real Local `autotitle:false`/idle conversation retained `title:null`, while eight browser batch GETs had 30.008–30.031-second intervals over 210 seconds. The direct adapter supplies a default display title. This does not exercise the Cloud null-title IDLE/FINISHED branch, which needs an approved Cloud conversation with a failed/unsupported title job. Run `backfill-18309-coordinator-checks`, Agent Server/SDK 1.54.0, automation 1.19.3; retained `evidence/F07.open-by-url/16877-local-polling.json` and `16877-local-idle.png`. Arrange the empty conversation with `api POST /api/conversations --write --data @<idle-autotitle-disabled-payload>`, open `/conversations/<id>`, clear `browser network`, leave the tab foreground, then read `browser network --filter ids= --last 100` and `conversation status <id>`. No model call or response was used.diff --git a/.agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md b/.agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.mdindex c91b49c5d2..a7e44cff20 100644--- a/.agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md+++ b/.agents/skills/verify-openhands/references/feature-map/F08-workspace-files-and-changes.md@@ -20,7 +20,7 @@ Source: `src/components/features/conversation/conversation-tabs/`, `src/componen- `F08.tabs-menu`: the ellipsis menu at the end of the tab bar lists every tab. Choosing a row opens that tab and closes the menu, and so does clicking outside.- `F08.tabs-pin`: the pill on each menu row pins or unpins a tab. Unpinning the active tab selects another pinned one, and the choice persists across reloads.- `F08.drawer-resize`: dragging the chat/drawer divider resizes both panes (chat 30–80 %, default 50 %, stored in `desktop-layout-panel-width`).-- `F08.vscode-link`: a `VSCode` button at the right of the tab bar opens the editor in a new window; it is hidden when the runtime has no editor. (#17660 removed it and #18048 restored it.)+- `F08.vscode-link`: a `VSCode` button at the right of the tab bar opens the editor in a new window; it is hidden when the runtime has no editor.- `F08.archived-disabled`: on an archived conversation the toggle is disabled with the tooltip `Unavailable for Archives`.- `F08.phone-panel-page`: at 1024 px or narrower the toggle navigates to `/conversations/<id>/panel`, a full-screen page with a Back chevron, the same tab bar and the tab body.- `F08.panel-direct-url`: opening `/conversations/<id>/panel` directly renders the panel page at 1024 px or narrower, and Back returns to the chat; wider, the URL is replaced with `/conversations/<id>` and the drawer opens beside the chat.@@ -102,6 +102,103 @@ Preconditions:- **Direct panel URL (`F08.panel-direct-url`).** At desktop run `control-openhands browser goto /conversations/<id>/panel`. The app replaces the URL with `/conversations/<id>` (the `goto` result already shows it; `control-openhands browser url` confirms), `control-openhands browser count 'testid=conversation-mobile-panel-back'` is `0`, `browser visible 'testid=chat-interface'` is `true` and `right-panel-toggle` `aria-pressed` is `true` (the drawer opens beside the chat; `browser screenshot --feature F08.panel-direct-url --name desktop-redirect`). Then run `control-openhands browser viewport phone` and the same `goto`: the URL keeps `/panel`, the `conversation-mobile-panel-back` count is `1` and `chat-interface` is not visible. `control-openhands browser click 'testid=conversation-mobile-panel-back' --expect-url '/conversations/[^/]+$'` returns to the chat. Return with `control-openhands browser viewport desktop`.- **Errors.** After each group run `control-openhands browser errors --app-only`. The Monaco disposal error above belongs to this family. 404s on `/api/llm/balance` and `.agents_tmp/` downloads come from the Usage and Planner tabs (F27).+- **Rich fenced code (`F08.files-rich-plain`).** Arrange a Markdown file in the selected workspace containing a heading and a fenced Python `print("qa")` block. Open it from Files in Rich mode and take `browser screenshot --feature F08.files-rich-plain --name fenced-code`. Known failure (reproduced 2026-10-12, [#16802](https://github.com/OpenHands/OpenHands/issues/16802)): the code renders in nested dark/gray containers with repeated padding. This is a layout-only observation; the measured `pre` and prose ancestor had zero border width, so it is not a claim of visible CSS borders.+- **Git filename with a space (`F08.uncommitted-changes`).** In a disposable workspace Git fixture, create and commit `my notes.md`, then modify it (plain shell Git/file commands are fixture setup, not UI proof). Open its empty Local conversation's Commits tab and read `browser network --filter '/git/'`. Expected: the modified path and other changes stay listed. Known failure (reproduced 2026-10-12, [OpenHands/software-agent-sdk#5642](https://github.com/OpenHands/software-agent-sdk/issues/5642)): the scoped changes call returns 400 and hides the Uncommitted changes. Only the space-name case was tested; the same issue's non-ASCII and rename cases still require their own checks.+- **File listing above one MiB (`F08.files-tree`).** In a disposable fixture, arrange 1800 tiny files beneath three 220-character directory segments so the sorted relative-path listing exceeds one MiB while remaining below the 2000-file cap. Keep a normal tracked README at the root. Open Files, click `testid=files-tab-refresh`, then count `role=button[name="README.md"][exact]` and inspect the tree. Known failure (reproduced 2026-10-12, [OpenHands/software-agent-sdk#5638](https://github.com/OpenHands/software-agent-sdk/issues/5638)): the README disappears and a spurious partial-path folder appears because the command API returns the tail output chunk only. The greater-than-100-chunk/null-exit case was not driven.++- **Subdirectory Git paths (`F08.uncommitted-changes`, `F08.diff-view-modes`).** Expected: a workspace at `repo/src` lists changes beneath that folder relative to it, and its diffs open. Known failure (reproduced 2026-10-16, SDK 1.55.0, [OpenHands/software-agent-sdk#5843](https://github.com/OpenHands/software-agent-sdk/issues/5843)): tracked paths retain the repository prefix while untracked paths are relative to `src`; opening displayed `src/app.py` requests `repo/src/src/app.py` and returns400. The ordinary repository-root control lists correct paths. The recipe creates only disposable, idle offline conversations.+ ```sh known-bug F08.diff-view-modes subdirectory-paths+ control-openhands browser viewport desktop+ QA_ID="$(python3 - <<'PY'+ import json, os, pathlib, subprocess, tempfile+ base = pathlib.Path(os.environ["OH_VERIFY_RUN"]) / "workspace"+ base.mkdir(exist_ok=True)+ fixture = pathlib.Path(tempfile.mkdtemp(prefix="qa-surface-", dir=base))+ llm = {"model":"openai/gpt-4o", "base_url":"http://127.0.0.1:1/v1", "api_key":"qa-not-real", "num_retries":0, "timeout":2}+ def create(workspace, parent=None, tags=None):+ payload = {"agent":{"kind":"Agent", "llm":llm, "tools":[]}, "workspace":{"working_dir":str(workspace)}, "autotitle":False}+ if parent: payload["parent_conversation_id"] = parent+ if tags: payload["tags"] = tags+ request = fixture / "request.json"+ request.write_text(json.dumps(payload))+ result = subprocess.run(["control-openhands", "api", "POST", "/api/conversations", "--write", "--data", "@"+str(request), "--pick", "id"], check=True, capture_output=True, text=True)+ return json.loads(result.stdout)["value"]+ def git(directory, *args):+ subprocess.run(["git", "-C", str(directory), *args], check=True, capture_output=True)+ repo = fixture / "repo"+ (repo / "src").mkdir(parents=True)+ git(repo, "init", "-q", "-b", "main")+ (repo / "README.md").write_text("root before\n")+ (repo / "src/app.py").write_text("x = 1\n")+ git(repo, "add", ".")+ git(repo, "-c", "user.name=QA", "-c", "[email protected]", "commit", "-qm", "Owned fixture")+ (repo / "README.md").write_text("root after\n")+ (repo / "src/app.py").write_text("x = 2\n")+ (repo / "src/new.py").write_text("new\n")+ print(create(repo / "src"))+ PY+ )"+ control-openhands browser goto "/conversations/$QA_ID"+ control-openhands browser click 'testid=right-panel-toggle'+ control-openhands browser click 'testid=conversation-tab-commits'+ control-openhands browser wait-text Uncommitted --timeout 10000+ control-openhands browser click 'testid=uncommitted-changes-row-toggle'+ control-openhands browser network --clear+ control-openhands browser click 'testid=file-diff-viewer-outer >> has-text=src/app.py >> testid=collapse' --observe 'testid=uncommitted-changes-row-content' --observe-ms 1000+ control-openhands browser screenshot --feature F08.diff-view-modes --name subdir-mixed-paths+ control-openhands assert --field recent.0.status --equals 200 --bug-shows 400 -- browser network --filter '/git/diff' # bug+ ```++- **Plain parent containing a repository (`F08.uncommitted-changes`).** Expected: a plain workspace holding one dirty repository exposes that repository's changes. Known failure (reproduced 2026-10-16, pinned SDK 1.55.0, [OpenHands/software-agent-sdk#5641](https://github.com/OpenHands/software-agent-sdk/issues/5641)): the changes request returns200 with no changes and Commits reads `No commits yet` although the child repository is dirty. The upstream fix is closed but not included in the current pin. The recipe removes only its own verified empty parent `.git`; never remove a user's repository.+ ```sh known-bug F08.uncommitted-changes nested-repository+ control-openhands browser viewport desktop+ QA_ID="$(python3 - <<'PY'+ import json, os, pathlib, subprocess, tempfile+ base = pathlib.Path(os.environ["OH_VERIFY_RUN"]) / "workspace"+ base.mkdir(exist_ok=True)+ fixture = pathlib.Path(tempfile.mkdtemp(prefix="qa-surface-", dir=base))+ llm = {"model":"openai/gpt-4o", "base_url":"http://127.0.0.1:1/v1", "api_key":"qa-not-real", "num_retries":0, "timeout":2}+ def create(workspace, parent=None, tags=None):+ payload = {"agent":{"kind":"Agent", "llm":llm, "tools":[]}, "workspace":{"working_dir":str(workspace)}, "autotitle":False}+ if parent: payload["parent_conversation_id"] = parent+ if tags: payload["tags"] = tags+ request = fixture / "request.json"+ request.write_text(json.dumps(payload))+ result = subprocess.run(["control-openhands", "api", "POST", "/api/conversations", "--write", "--data", "@"+str(request), "--pick", "id"], check=True, capture_output=True, text=True)+ return json.loads(result.stdout)["value"]+ def git(directory, *args):+ subprocess.run(["git", "-C", str(directory), *args], check=True, capture_output=True)+ import shutil+ parent = fixture / "parent"+ child = parent / "proj"+ child.mkdir(parents=True)+ git(child, "init", "-q", "-b", "main")+ (child / "README.md").write_text("before\n")+ git(child, "add", ".")+ git(child, "-c", "user.name=QA", "-c", "[email protected]", "commit", "-qm", "Owned nested fixture")+ (child / "README.md").write_text("after\n")+ (child / "notes.txt").write_text("untracked\n")+ conversation_id = create(parent)+ empty_git = parent / ".git"+ if empty_git.exists():+ assert empty_git.is_dir() and not empty_git.is_symlink()+ top = pathlib.Path(subprocess.check_output(["git", "-C", str(parent), "rev-parse", "--show-toplevel"], text=True).strip())+ assert top.resolve() == parent.resolve()+ assert not subprocess.check_output(["git", "-C", str(parent), "ls-files"], text=True).strip()+ shutil.rmtree(empty_git)+ print(conversation_id)+ PY+ )"+ control-openhands browser goto "/conversations/$QA_ID"+ control-openhands browser click 'testid=right-panel-toggle'+ control-openhands browser click 'testid=conversation-tab-commits'+ control-openhands browser wait '#conversation-tab-panel main'+ control-openhands browser wait 'testid=commits-tab-status' --state detached --timeout 10000+ control-openhands assert --field recent.0.status --equals 200 -- browser network --filter '/git/changes'+ control-openhands browser screenshot --feature F08.uncommitted-changes --name nested-repository+ control-openhands assert --field count --equals 1 --bug-shows 0 -- browser count 'testid=uncommitted-changes-row' # bug+ ```+## Gotchas- When the drawer is closed, its tabs stay mounted at width 0. `browser count 'testid=files-tab'` is `1` even then, so assert with `browser visible` or `right-panel-toggle` `aria-pressed`.@@ -119,3 +216,18 @@ Preconditions:- At phone width, chat path links (and the agent's `canvas_ui_control` tab and file requests, through `src/services/canvas-ui.ts`) navigate to `/conversations/<id>/panel` (#18045); at desktop width they open the side drawer and the URL does not change. A `/panel` URL above 1024 px redirects to `/conversations/<id>`, so a bullet that leaves the phone viewport on `/panel` ends on the chat route.- If the agent's terminal resets mid-command ("terminal session was reset", for example after a `restart`), the agent reruns the command. Check the workspace with `git status` through a second prompt rather than trusting the first observation. Runs no longer share a tmux server (OpenHands/OpenHands#17946).- Known issue: [#17567](https://github.com/OpenHands/OpenHands/issues/17567) (Monaco disposed-model page error on diff unmount; it also fires on collapse and on a diff→old switch).+- Known issue [#18108](https://github.com/OpenHands/OpenHands/issues/18108) (`F08.files-tree`): Files tab shows "No files in workspace" on cloud when the agent writes outside /workspace/project. Live prerequisite: An authenticated Enterprise Cloud sandbox, no repository selected, with files outside its default working directory.+- Known issue [#17773](https://github.com/OpenHands/OpenHands/issues/17773) (`F08.files-tree`): Files tab always shows "No files in workspace" on Windows local backend. Live prerequisite: A Windows local Agent Server using cmd.exe; this run is macOS.+- Known issue [#16949](https://github.com/OpenHands/OpenHands/issues/16949) (`F08.files-refresh`): Keep Files and Changes views source-driven after refresh. Live prerequisite: An authenticated Cloud repository for the refresh-query mismatch, or real file-editor observations causing an in-place M-to-D Git refetch.+- Known issue [#16716](https://github.com/OpenHands/OpenHands/issues/16716) (`F08.uncommitted-changes`): Diff view is regularly out of sync. Live prerequisite: The original simultaneous-agent/shared-repository state or a deterministic stale-diff trigger; the report only suggests that timing.+- Known issue [#16689](https://github.com/OpenHands/OpenHands/issues/16689) (`F08.files-load-error`): When `workspace/` paths throw a 401, call the workspace-session endpoint. Live prerequisite: The original fresh-VM URL, auth/cookie state and exact request sequence; the report says the trigger is unclear.+- Known issue [#16454](https://github.com/OpenHands/OpenHands/issues/16454) (`F08.uncommitted-changes`): Diff list is showing duplicate entries. Live prerequisite: The original Git changes response and update sequence producing duplicate rows; the report provides no trigger.+- Known issue [#16331](https://github.com/OpenHands/OpenHands/issues/16331) (`F08.files-tree`): file preview panel stops working during streaming. Live prerequisite: A real streaming/tool-running conversation while selecting a second workspace file.+- Known issue [#15692](https://github.com/OpenHands/OpenHands/issues/15692) (`F08.uncommitted-changes`): Diff still shows after agent has committed and pushed. Live prerequisite: A real agent commit/push terminal observation while the diff remains mounted, plus a disposable Git remote.+- Known issue [#15567](https://github.com/OpenHands/OpenHands/issues/15567) (`F08.files-load-error`): Workspace file preview returns 401 Unauthorized on non-HTTPS non-loopback hosts. Live prerequisite: A plain-HTTP non-loopback deployment and browser accessing that origin, with workspace-cookie observation.+- Known issue [#15519](https://github.com/OpenHands/OpenHands/issues/15519) (`F08.files-load-error`): "files" tab in Files side panel throws 401 Unauthorized. Live prerequisite: The reported Linux Docker deployment/auth topology; this isolated run uses the macOS source launcher.+- Known issue [#15501](https://github.com/OpenHands/OpenHands/issues/15501) (`F08.files-rich-plain`): Generated-page JavaScript was blocked by an iframe sandbox. Files Rich intentionally disables scripts; the report does not identify whether Files or Browser was used. Live prerequisite: The original preview panel and file URL or minimal page identifying the intended surface; Files Rich deliberately blocks scripts, while Browser is separate.+- Known issue [OpenHands/software-agent-sdk#4562](https://github.com/OpenHands/software-agent-sdk/issues/4562) (`F08.files-load-error`): Plain-HTTP non-loopback workspace previews can return401 after the browser rejects the workspace-session cookie. Live prerequisite: A plain-HTTP non-loopback same-origin browser deployment with real workspace-cookie observation.+- Known issue [#16871](https://github.com/OpenHands/OpenHands/issues/16871) (`F08.vscode-link`): the shared `transformVSCodeUrl` still accepts arbitrary schemes, and the current Cloud `use-unified-vscode-url.ts` branch passes sandbox `exposed_urls` editor values through it without the Local origin gate. Live prerequisite: an authorized Cloud account/sandbox whose editor `exposed_urls` value can be controlled for safe URL-scheme rejection checks. Local inspection on 2026-10-12 at `fa1f724ab573` found no Download via VS Code action in the real card menu: its only production caller does not supply the required `showOptions` prop. The current drawer opens with `noopener,noreferrer`. Neither the old opener exploit nor the remaining Cloud scheme scenario was reproduced. Evidence: `backfill-18309-surface-rechecks/F04.download/vscode-card-action-absent-16871.png` and its ARIA snapshot. Keep this linked to the live editor-link consumer; do not infer the whole issue resolved from the removed card action.+- Known issue [OpenHands/software-agent-sdk#5845](https://github.com/OpenHands/software-agent-sdk/issues/5845) (`F08.vscode-link`): reserved characters in the workspace path can split the editor URL's unencoded `folder` parameter and open the wrong folder. Live prerequisite: a supported origin serving a genuinely enabled/running editor and an owned conversation in a path containing `&` or `#`. A direct URL-constructor call does not exercise the link; Cloud `exposed_urls` takes another path.+- Known issue [OpenHands/software-agent-sdk#5784](https://github.com/OpenHands/software-agent-sdk/issues/5784) (`F08.files-rich-plain`, `F08.files-fallbacks`, `F08.files-open-new-window`): a path-prefixed Agent Server can set a workspace cookie whose Path omits the prefix, leaving HTML/image/file previews unauthorized. Live prerequisite: an authenticated path-prefixed owned Local server with `OH_WEB_URL` and a real cookie-authenticated preview; the ordinary root-path launch cannot exercise that cookie Path.diff --git a/.agents/skills/verify-openhands/references/feature-map/F09-settings-shell.md b/.agents/skills/verify-openhands/references/feature-map/F09-settings-shell.mdindex b2e83f335b..ac15a55a28 100644--- a/.agents/skills/verify-openhands/references/feature-map/F09-settings-shell.md+++ b/.agents/skills/verify-openhands/references/feature-map/F09-settings-shell.md@@ -95,3 +95,5 @@ Preconditions:- From 768 to 1023 px the sidebar rail stays and the phone top bar is gone; the Settings selector above each subpage provides sibling navigation. The rail's gear (`backend-selector-settings-link`) still returns to the hub.- The phone hub is indented 32 px while sub-pages start at 16 px (hub padding inside the page padding); that is current layout, not overflow.- The "LLM Profiles" rename in `useSettingsNavItems` targets a `/settings` nav item that does not exist, so it never applies; the LLM nav item always reads "LLM".+- Known issue [#17508](https://github.com/OpenHands/OpenHands/issues/17508) (`F09.update-available`): After installing a newer package, an older running bundle can keep showing its own version and an update banner. Live prerequisite: Isolated old/new package installs with the old process still serving, or the original stale bundle/cache capture after update.+- The public `SettingsNavigation` export renders the legacy `SettingsMobileDrawer` with a half-opacity black scrim (`bg-black/50`). Check its modern and fallback compiled CSS as a build contract. The routed phone hub uses `SettingsMobileHub`, so its navigation recipe does not exercise the exported drawer; visual verification of that export needs an embedding host.diff --git a/.agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md b/.agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.mdindex 7a055960ae..6ba2696988 100644--- a/.agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md+++ b/.agents/skills/verify-openhands/references/feature-map/F10-llm-profiles.md@@ -92,7 +92,7 @@ Preconditions:- **Link to a provider connection (`F10.connection-link`, `F10.list-grouped`).** Arrange a connection through the F11 form: `control-openhands browser click 'testid=add-provider-connection'`, `control-openhands browser fill 'testid=provider-connection-name-input' QA_ds_conn`, `control-openhands browser click 'testid=provider-connection-provider-input'`, `control-openhands browser fill 'testid=provider-connection-provider-input' deepseek`, `control-openhands browser click 'testid=provider-item-deepseek'`, `control-openhands browser fill 'testid=provider-connection-api-key-input' --value-file <key file>`, `control-openhands browser click 'testid=provider-connection-submit'` and `control-openhands browser wait-text 'Connection "QA_ds_conn" created'`. Edit `QA_adv`, click `testid=sdk-section-basic-toggle`, run `control-openhands browser click 'testid=llm-provider-connection-input'` (options `None`, `QA_ds_conn`) and `control-openhands browser click 'role=option[name="QA_ds_conn"]'`; `browser count 'testid=llm-api-key-input'` is `0`, and on the Advanced tab `browser count 'testid=base-url-input'` is `0`. Picking `role=option[name="None"]` brings the key field back (`1`). Re-select `QA_ds_conn`, then `control-openhands browser click 'testid=save-profile-btn' --observe 'testid=save-profile-btn' --observe-ms 3000`: states `Save`, `Saving...` (no `Validating...`), and `wait-text 'Profile "QA_adv" updated'`. After a reload `control-openhands browser eval "[...document.querySelectorAll('[data-testid=profile-group-header]')].map(e=>e.innerText).join(' | ')"` is `QA_DS_CONN | NOT LINKED` (CSS upper-case) and `control-openhands browser count 'div:has(> [data-testid=profile-group-header]:text-is("QA_ds_conn")) [data-testid=profile-row]:has([title="QA_adv"])'` is `1` (screenshot `--feature F10.list-grouped --name grouped`).- **Add models from a provider (`F10.add-models`).** Run `control-openhands browser click 'testid=add-models-from-provider'`; `browser enabled 'testid=add-models-submit'` is `false`. Keyless: `control-openhands browser select 'testid=add-models-provider' moonshot` and `control-openhands browser wait 'testid=add-models-select-all'`; `browser text 'testid=add-models-keyless-note'` is `No connection — profiles will need a key added later`, rows start unchecked (`Add 0 profiles`), and `control-openhands browser check 'testid=add-models-select-all'` makes `browser text 'testid=add-models-submit'` read `Add 12 profiles`; uncheck it again. Linked: `control-openhands browser select 'testid=add-models-provider' deepseek`; the connection field (`testid=add-models-connection-field`, options `No connection (keyless)` and `QA_ds_conn`) is pre-bound to `QA_ds_conn`. Run `control-openhands browser check 'testid=add-models-check-deepseek/deepseek-chat'`, `control-openhands browser check 'testid=add-models-check-deepseek/deepseek-v3'`, `control-openhands browser click 'testid=add-models-submit'` and `control-openhands browser wait-text 'Added 2'` (toast `Added 2 profiles`). The modal closes; after a reload `deepseek-chat` and `deepseek-v3` sit under the `QA_DS_CONN` header and `api GET /api/profiles` shows both with a `provider_connection_id`. Preselect entry: `control-openhands browser click 'testid=provider-connection-row >> has-text=QA_ds_conn >> testid=provider-connection-menu-trigger'` (menu: Bulk add, Edit, Delete), `control-openhands browser click 'testid=provider-connection-actions-menu >> testid=provider-connection-add-models'`; provider `deepseek` and the connection are preset, every remaining row is checked (`Add 10 profiles`), and `browser count 'testid=add-models-row-deepseek/deepseek-chat'` is `0` (already added). Close with `control-openhands browser click 'role=dialog[name="Add models as profiles"] >> role=button[name="Cancel"]'`.- **Broken link (`F10.broken-link`).** The local server refuses to delete a referenced connection (row menu → `testid=provider-connection-delete` → `testid=delete-provider-connection-confirm` toasts `Provider connection cannot be deleted while it is referenced by LLM profile(s): ...`), so arrange the state directly: `control-openhands api POST /api/profiles/QA_broken --write --data '{"llm":{"model":"deepseek/deepseek-chat","provider_connection_id":"00000000000000000000000000000000"},"include_secrets":false}'`. Then `control-openhands browser reload` and `control-openhands browser text '[data-testid=profile-row]:has([title="QA_broken"])'` reads `QA_broken deepseek/deepseek-chat Broken link`; `control-openhands browser attr '[data-testid=profile-row]:has([title="QA_broken"]) >> testid=profile-broken-connection-badge' title` is `The linked provider connection was deleted. Edit this profile to re-link it.` (screenshot `--feature F10.broken-link --name badge`). Edit it: `browser value 'testid=llm-provider-connection-input'` is the orphaned id and the listbox offers `None`, `QA_ds_conn` and that id. The row's group header shows the raw id, not a name (see Gotchas).-- **ChatGPT subscription, pre-auth (`F10.subscription`).** In a fresh editor run `control-openhands browser click 'testid=sdk-section-basic-toggle'` first (after the schema step the Default profile has a temperature, so a new editor opens on All; see `F10.create-prefill`), then `control-openhands browser click 'testid=llm-auth-type-input'` (options `API key`, `ChatGPT subscription`) and `control-openhands browser click 'role=option[name="ChatGPT subscription"]'`. `browser text 'testid=subscription-status'` is `ChatGPT subscription not connected`, `browser value 'testid=llm-subscription-model-input'` is `gpt-5.5` (list: gpt-5.5, gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra, gpt-6-astra), and `llm-provider-input` and `llm-api-key-input` are gone (count `0`). Known failure (reproduced 2026-10-13, #17135): run `control-openhands browser fill 'testid=profile-name-input' QA_sub`, `control-openhands browser network --clear`, `control-openhands browser click 'testid=save-profile-btn'` and `control-openhands browser network --bodies --filter '/api/profiles'`. Expected: Save validates and saves the shown `gpt-5.5`. Actual: one `POST /api/profiles/QA_sub/validate` whose body has `"model":"openai/gpt-5.6-sol"`, the API-key form's hidden default, although the dropdown shows `gpt-5.5` (it shows the first subscription model without writing it to the form); `browser toasts --history` has `OpenAI subscription login is required`, because no subscription is connected. Before #18218 (BASE 553d4519, twice) the same Save toasted `Model is required` and sent nothing. Picking a model in `testid=llm-subscription-model-input`, even `gpt-5.5` again, makes the next Save send `POST /api/profiles/QA_sub/validate`, which answers `OpenAI subscription login is required` here (`browser toasts --history`), because no subscription is connected. `control-openhands browser click 'testid=subscription-connect'` and `control-openhands browser wait 'testid=subscription-user-code'`: the card shows the code, `Open sign-in page`, `I've finished signing in` (`subscription-poll`) and Cancel, and `control-openhands browser tabs` lists a second page on `auth.openai.com` (it opens a second or two after the code appears; rerun `browser tabs` if it lists one page). The card polls on its own, so `Sign-in is not complete yet. ...` can already show before you click poll. Close it with `control-openhands browser close-tab 1` and `control-openhands browser tab 0`. `control-openhands browser click 'testid=subscription-poll'` leaves `Sign-in is not complete yet. Finish the browser flow and try again.` on the card, and `control-openhands browser click 'testid=subscription-cancel'` removes the code (`count 'testid=subscription-user-code'` `0`). Switching back with `role=option[name="API key"]` restores provider `OpenAI` and model `gpt-5.6-sol`. Completing sign-in is blocked (needs a ChatGPT Plus/Pro account).+- **ChatGPT subscription, pre-auth (`F10.subscription`).** In a fresh editor run `control-openhands browser click 'testid=sdk-section-basic-toggle'` first (after the schema step the Default profile has a temperature, so a new editor opens on All; see `F10.create-prefill`), then `control-openhands browser click 'testid=llm-auth-type-input'` (options `API key`, `ChatGPT subscription`) and `control-openhands browser click 'role=option[name="ChatGPT subscription"]'`. `browser text 'testid=subscription-status'` is `ChatGPT subscription not connected`, `browser value 'testid=llm-subscription-model-input'` is `gpt-5.5` (list: gpt-5.5, gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra, gpt-6-astra), and `llm-provider-input` and `llm-api-key-input` are gone (count `0`). Known failure (reproduced 2026-10-13, #17135): run `control-openhands browser fill 'testid=profile-name-input' QA_sub`, `control-openhands browser network --clear`, `control-openhands browser click 'testid=save-profile-btn'` and `control-openhands browser network --bodies --filter '/api/profiles'`. Expected: Save validates and saves the shown `gpt-5.5`. Actual: one `POST /api/profiles/QA_sub/validate` whose body has `"model":"openai/gpt-5.6-sol"`, the API-key form's hidden default, although the dropdown shows `gpt-5.5` (it shows the first subscription model without writing it to the form); `browser toasts --history` has `OpenAI subscription login is required`, because no subscription is connected. Picking a model in `testid=llm-subscription-model-input`, even `gpt-5.5` again, makes the next Save send `POST /api/profiles/QA_sub/validate`, which answers `OpenAI subscription login is required` here (`browser toasts --history`), because no subscription is connected. `control-openhands browser click 'testid=subscription-connect'` and `control-openhands browser wait 'testid=subscription-user-code'`: the card shows the code, `Open sign-in page`, `I've finished signing in` (`subscription-poll`) and Cancel, and `control-openhands browser tabs` lists a second page on `auth.openai.com` (it opens a second or two after the code appears; rerun `browser tabs` if it lists one page). The card polls on its own, so `Sign-in is not complete yet. ...` can already show before you click poll. Close it with `control-openhands browser close-tab 1` and `control-openhands browser tab 0`. `control-openhands browser click 'testid=subscription-poll'` leaves `Sign-in is not complete yet. Finish the browser flow and try again.` on the card, and `control-openhands browser click 'testid=subscription-cancel'` removes the code (`count 'testid=subscription-user-code'` `0`). Switching back with `role=option[name="API key"]` restores provider `OpenAI` and model `gpt-5.6-sol`. Completing sign-in is blocked (needs a ChatGPT Plus/Pro account).- **Delete (`F10.delete`, `F10.delete-default`).** Run `control-openhands browser click '[data-testid=profile-row]:has([title="QA_pro-copy"]) >> testid=profile-menu-trigger'` and `control-openhands browser click 'testid=profile-actions-menu >> testid=profile-delete'`. `control-openhands browser snapshot 'role=dialog'` shows `Delete Profile` with `Are you sure you want to delete the profile "QA_pro-copy"? This action cannot be undone.`, and `browser eval "document.activeElement.innerText"` is `Cancel`. `control-openhands browser click 'role=dialog[name="Delete Profile"] >> role=button[name="Cancel"]'` keeps the row (count `1`). Repeat, then `control-openhands browser click 'testid=delete-profile-confirm'`, `control-openhands browser wait-text 'Profile "QA_pro-copy" deleted'`, reload, and the count is `0`. Deleting the Default row the same way leaves exactly one `profile-active-badge` on another profile (`browser text '[data-testid=profile-row]:has([data-testid=profile-active-badge])'`), and `api GET /api/profiles` names it in `active_profile`.- **Command menu and direct URL (`F10.list`).** Run `control-openhands browser goto /` (the bullets above end on `/settings/llm`, where the URL wait would pass at once), `control-openhands browser press Control+k`, `control-openhands browser type 'testid=command-menu >> role=combobox' LLM` and `control-openhands browser snapshot 'testid=command-menu'`: two options, `LLM profiles` ("Manage models, providers, and API keys.") and `Model Router` (F12; its keywords include "llm"). Hovering an option selects it, so which one Enter runs depends on where the mouse rests: click the option instead, `control-openhands browser click 'testid=command-menu >> role=option[name="LLM profiles"]' --expect-url '/settings/llm(\?|$)'`. `browser count 'testid=add-llm-profile'` is `1` on arrival. `control-openhands browser goto /settings/llm` shows the same list (`browser count 'testid=profile-row'`, one `profile-active-badge`; screenshot `--feature F10.list --name list`).- **Phone layout (`F10.phone`).** Run `control-openhands browser viewport phone`, `control-openhands browser goto /settings/llm`, `control-openhands browser bbox 'testid=add-llm-profile'`, `control-openhands browser bbox 'testid=profile-row >> nth=0'` and `control-openhands browser screenshot --feature F10.phone --name list`: `insideViewport` `true`, `pageHorizontalOverflow` `false`. Open a row menu: `browser bbox 'testid=profile-actions-menu'` is inside the viewport. Open Edit: `browser bbox 'testid=profile-name-input'` is 348 px wide inside the 390 px viewport and `pageHorizontalOverflow` is `false` (`llm-settings-screen` may report `insideViewport` `false` when the edited profile's form is taller than the screen; only the overflow flag matters). Return with `control-openhands browser click 'testid=cancel-profile-btn'` and `control-openhands browser viewport desktop`.@@ -101,6 +101,51 @@ Preconditions:- **Load error (`F10.list-states`).** With any profile linked to a connection (here `QA_adv` → `QA_ds_conn`, from the Link bullet), make this run's connections store unreadable (shell arrange, this run only): `F=$OH_VERIFY_RUN/private/provider-connections/provider_connections.json; cp $F $F.bak; printf '{not json' > $F`. `control-openhands api GET /api/profiles` is now `400` (`Provider connections file is unreadable: …`). Run `control-openhands browser reload`, `control-openhands browser wait 'testid=profiles-load-error'` and `control-openhands browser wait 'testid=provider-connections-load-error'` (it can appear one retry after the profiles error): the first reads `Failed to load profiles` in the theme's error colour (`browser eval "getComputedStyle(document.querySelector('[data-testid=profiles-load-error] p')).color"` is `rgb(231, 106, 94)` on OpenHands-Neutral and `rgb(192, 40, 45)` on Light+, #18330), with both Add buttons still shown and `provider-connections-load-error` below it (`F11.load-error`); screenshot `--feature F10.list-states --name load-error-neutral`. Restore with `cp $F.bak $F && rm $F.bak` and `browser reload`. The backend-down route does not reach it: `control-openhands service stop agent-server` and `control-openhands browser reload` do not reach `testid=profiles-load-error`: the whole app is replaced by the disconnected-backend screen (`testid=agent-server-onboarding-screen`, toasts `An error occurred` (twice, one after the other) and `Could not connect to the configured agent server. Make sure it is running and reachable, then reload the page.` (#18051); the Manage backends row detail still reads `HTTP request failed (502 Bad Gateway): "Bad Gateway: connect ECONNREFUSED 127.0.0.1:<agent-server port>"`). The list error needs `/api/profiles` alone to fail. Bring the stack back with `control-openhands restart`.- **Cleanup.** Run `control-openhands browser goto /settings/llm` (the Load error bullet left the disconnected-backend screen), then delete every `QA_*`, `deepseek-chat` and `deepseek-v3` profile through the row menu, linked ones first; `control-openhands browser count 'testid=profile-group-header'` drops to `0` once no linked row is left (`F10.list-grouped`). Then delete `QA_ds_conn` (`testid=provider-connection-row >> has-text=QA_ds_conn >> testid=provider-connection-menu-trigger`, `testid=provider-connection-actions-menu >> testid=provider-connection-delete`, `testid=delete-provider-connection-confirm`; toast `Connection "QA_ds_conn" deleted`). With no profiles left, `testid=profiles-empty` returns and `api GET /api/profiles` is `{"profiles":[],"active_profile":null}`.+- **Embedded editor starts on Basic (`F10.create-prefill`, `F10.edit`).** Arrange an active disposable `QA_LOOPBACK` profile through the real profiles API with `llm.model` `openai/gpt-4o`, a dummy key, `llm.base_url` `http://127.0.0.1:1/v1`, `num_retries` `0` and temperature `0.4` (no completion). Run `control-openhands browser goto /settings/llm`, `control-openhands browser click 'testid=add-llm-profile'` and `control-openhands browser snapshot 'role=tablist'`. Expected by [#17801](https://github.com/OpenHands/OpenHands/issues/17801): Basic selected each time the embedded form opens. Known failure, reproduced 2026-10-12: All is selected. Back, open the exact `QA_LOOPBACK` row menu and `testid=profile-actions-menu >> testid=profile-edit`; the tablist also selects All. Repeat Add/Edit once, then Back without saving. This exercises the non-default settings condition; it does not claim that a default-only profile opens on All.++- **Off-catalog Basic model (`F10.model-picker`).** From the list open Add, click `testid=sdk-section-basic-toggle`, then `control-openhands browser choose 'testid=llm-provider-input' OpenRouter`. Run `control-openhands browser fill 'testid=llm-model-input' 'thinkingmachines/inkling-small:free'`, `control-openhands browser press Tab` and `control-openhands browser value 'testid=llm-model-input'`; repeat fill/Tab once. Expected: keep the model entered under the selected provider. Known failure, reproduced 2026-10-12: the value is empty on both attempts ([#17577](https://github.com/OpenHands/OpenHands/issues/17577), concrete companion of [#17233](https://github.com/OpenHands/OpenHands/issues/17233)). Capture `browser screenshot --feature F10.model-picker --name off-catalog-discarded` and Back without saving. The Advanced provider-prefix/authenticated-save half needs a working OpenRouter route; this check proves the Basic entry dead end.++- **Azure Base URL on Basic (`F10.model-picker`).** Open Add, explicitly select Basic and `control-openhands browser choose 'testid=llm-provider-input' Azure`. `control-openhands browser count 'testid=base-url-input'` should be `1` for Azure's required endpoint. Known failure, reproduced 2026-10-12 ([#16726](https://github.com/OpenHands/OpenHands/issues/16726)): Basic count `0`, after `browser click 'testid=sdk-section-advanced-toggle'` count `1`, and after `browser click 'testid=sdk-section-basic-toggle'` count `0` again. Screenshot Basic and Back without saving. No Azure account is needed for this form omission; its reported authenticated conversation error is outside this observation.++- **Slow invalid preflight (`F10.create-validation`).** Arrange the disposable dummy-key `QA_LOOPBACK` profile from the embedded-editor bullet (all model traffic to `http://127.0.0.1:1/v1`). Open Add, explicitly select All and fill `testid=sdk-settings-llm.num_retries` with `5`; select Advanced, fill `llm-custom-model-input` with `openai/gpt-4o`, `base-url-input` with the refused loopback URL, the key via `--value-file <dummy-key file>` and `profile-name-input` with `QA_SLOW`. Clear `browser network --filter QA_SLOW --clear`, click Save with `--observe 'testid=save-profile-btn' --observe-ms 1000`, then inspect toasts and network after the 30-second client deadline. Expected: refuse the unreachable endpoint and save nothing. Known failure, reproduced twice 2026-10-12 ([#15778](https://github.com/OpenHands/OpenHands/issues/15778), slow branch of [OpenHands/software-agent-sdk#5100](https://github.com/OpenHands/software-agent-sdk/issues/5100)): `Validating...` lasts until the browser aborts; 30 seconds after the validation POST, a profile POST succeeds with 201 and the profile appears in the list. The frontend deliberately returns no verdict for an abort, then saves. Repeating under `QA_SLOW_2` behaves the same; capture the list and `browser network --filter QA_SLOW_2 --bodies`. With retries `1`, the same refused endpoint instead failed in under a second, showed `Connection error` and left `GET /api/profiles/QA_INVALID` at 404 on both saves. Delete the two saved QA_SLOW profiles via the real API as fixture cleanup. This does not verify [OpenHands/software-agent-sdk#5100](https://github.com/OpenHands/software-agent-sdk/issues/5100)'s separate exhausted-budget 429 branch, which needs an approved real route returning that provider verdict.++- **Deleted title profile after failed cleanup (`F10.delete`).** Arrange a non-default dummy loopback profile `QA_Titles3`, select it on Application's Title generation model dropdown and Save there (this also hydrates the frontend settings cache). Navigate back via `sidebar-settings-/settings/llm`, open the exact profile row menu and Delete. On macOS arrange a selective real write failure by recording `settings.json`'s original file flags, setting only `$OH_VERIFY_RUN/private/settings.json` to `stat.UF_IMMUTABLE` with `os.chflags`, and restoring its flags in a `finally` block immediately after the delete observation. Child `profiles/` stays writable and settings stay readable; ordinary chmod is insufficient because SDK Save repairs directory permissions. Clear the network log, confirm Delete, read toasts, the exact row count and `api GET /api/profiles/QA_Titles3`. Expected: remove the row despite the cleanup error. Known failure, reproduced twice 2026-10-12 ([#17520](https://github.com/OpenHands/OpenHands/issues/17520)): DELETE returns 200, the following title-cleanup PATCH returns 500 on three retries, the error surfaces, but the row remains (`1`) while the API returns 404. Repeat with `QA_Titles4`. Restore flags before `doctor`; clear the stale title preference through the real settings API. No provider validation or network interception is needed.+++- **Connection-linked provider dispatch (`F10.custom-model`, `F10.connection-link`).** From `/settings/llm` Add > Advanced, save `QA_QWEN` with `llm-custom-model-input` `qwen` linked through `llm-provider-connection-input` to an own OpenAI connection at refused loopback. Activate it and start two real Home chats; inspect `conversation events <id> --kinds ConversationErrorEvent --last 2`. Expected: the chosen OpenAI-compatible connection supplies the transport. Known failure, reproduced twice 2026-10-12 ([#17261](https://github.com/OpenHands/OpenHands/issues/17261)): Save succeeds, but each actual chat fails before HTTP with `LLM Provider NOT provided` for `model=qwen`. The reported Windows installation, unlinked preflight and successful local inference are further branches. With a separate own connection whose provider is `custom_openai`, repeat the UI Save/start for `ollama-local/qwen`: the same failure reproduces twice for the namespace reported in [#17195](https://github.com/OpenHands/OpenHands/issues/17195). Its exact exposed model suffix is absent from the report, so `qwen` is an owned diagnostic suffix; this is not a complete 9Router/Ollama inference check. Restore the neutral active profile and remove only the own profiles/connections.++- **Bedrock named-profile import (`F10.default-used`).** In the run-owned launcher environment set `AWS_EC2_METADATA_DISABLED=true`, then `restart` and `doctor`; verify that flag without printing other environment contents. Arrange an own profile with `model: bedrock/anthropic.claude-3-haiku-20240307-v1:0`, `aws_profile_name: qa_nonexistent`, region `us-east-1`, retries `0`, and both `base_url` and `aws_bedrock_runtime_endpoint` `http://127.0.0.1:1/v1`. A dummy-only API-key presence marker is needed for Canvas's send gate; the SDK filters that non-Bedrock key on this IAM path. Activate and start a real Home chat; inspect the error privately. Expected: the dependency imports and the missing named profile is reported before any AWS/provider request. A connection error cannot prove successful Bedrock inference. Restore the launcher environment and active profile, then remove only the own fixture.++- **Unavailable subscription upstream is handled (`F10.subscription`).** Expected: device-start failure reports502/504 rather than a server crash. Known failure (reproduced twice 2026-10-16, [OpenHands/software-agent-sdk#5880](https://github.com/OpenHands/software-agent-sdk/issues/5880)): real Connect returned500 with the backend offline. The block uses an owned offline proxy; browser offline alone does not cut the backend request. The cached runtime must already exist: `UV_OFFLINE=1` prevents its dependency resolver from reaching the blocked network. The block changes only this run's environment, keeps model defaults on refused loopback, and restores/restarts it in `EXIT`. The poll branch remains blocked without a genuine pending device token.++ ```sh known-bug F10.subscription offline-start+ QA_SUB_ENV="$OH_VERIFY_RUN/private/qa-known-subscription-env.json"+ cp "$OH_VERIFY_RUN/private/launcher-env.json" "$QA_SUB_ENV"+ trap 'cp "$QA_SUB_ENV" "$OH_VERIFY_RUN/private/launcher-env.json"; rm -f "$QA_SUB_ENV"; control-openhands restart --timeout 120 >/dev/null; control-openhands browser goto /settings/llm >/dev/null' EXIT+ python3 - <<'PYENV'+ from pathlib import Path+ import os,json+ p=Path(os.environ['OH_VERIFY_RUN'])/'private/launcher-env.json'+ d=json.loads(p.read_text())+ assert d.get('OPENAI_BASE_URL')=='http://127.0.0.1:1/v1'+ assert d.get('OPENAI_API_BASE')=='http://127.0.0.1:1/v1'+ d['UV_OFFLINE']='1'+ for key in ['HTTP_PROXY','HTTPS_PROXY','ALL_PROXY','http_proxy','https_proxy','all_proxy']: d[key]='http://127.0.0.1:1'+ d['NO_PROXY']=d['no_proxy']='127.0.0.1,localhost,::1'+ p.write_text(json.dumps(d))+ PYENV+ control-openhands restart --timeout 120+ control-openhands doctor+ control-openhands browser viewport desktop+ control-openhands browser goto /settings/llm+ control-openhands browser click 'testid=add-llm-profile'+ control-openhands browser click 'testid=sdk-section-basic-toggle'+ control-openhands browser choose 'testid=llm-auth-type-input' 'ChatGPT subscription'+ control-openhands browser network --clear+ control-openhands browser click 'testid=subscription-connect' --observe 'role=status' --observe-ms 1000+ control-openhands assert --field recent.0.status --matches '^(502|504)$' --bug 'recent.0.status eq 500 && recent.0.path eq "/api/llm/subscription/openai/device/start"' -- browser network --filter '/api/llm/subscription/openai/device/start' # bug+ ```+## Gotchas- Save always runs a real completion against the provider (except for connection-linked profiles). A bad or missing key fails fast with the provider's message; budget one tiny call per Save.@@ -120,4 +165,17 @@ Preconditions:- The list follows the server's order, which sorts by profile file name: `QA_pro-copy-1` lists before `QA_pro-copy` (`-` sorts before `.json`'s `.`). Do not assert a strict alphabetical order on names that are prefixes of each other.- Toasts last only a few seconds. A `browser screenshot` taken after a slow command can miss them, so assert with `browser wait-text` or `browser toasts`.- A new profile keeps the inherited LLM options the editor shows, such as temperature.-- `Subscription models are not loaded yet.` is translated in every language (`SETTINGS$SUBSCRIPTION_MODELS_NOT_LOADED`, #18308; source check 2026-10-13, not driven live). It shows when a subscription profile is saved before `GET /api/llm/subscription/openai/models` answers, in the settings save (`src/routes/llm-settings.tsx`) and in onboarding's LLM step. No recipe drives it, since it needs that request to be slow or to fail.+- `Subscription models are not loaded yet.` uses the translated `SETTINGS$SUBSCRIPTION_MODELS_NOT_LOADED` copy in settings and onboarding. It shows when a subscription profile is saved before `GET /api/llm/subscription/openai/models` answers; checking that branch needs the real request to be pending or failed.+- Known issue [#17785](https://github.com/OpenHands/OpenHands/issues/17785): Saving DeepSeek-v4-flash on Windows Desktop can fail with missing cache_creation_tokens instead of saving. The reported path needs a Windows Desktop 1.24.0 environment with its SDK dependency resolution and an approved DeepSeek key; this macOS source run cannot establish that result. (`F10.create-validation`).+- Known issue [#17706](https://github.com/OpenHands/OpenHands/issues/17706): A freshly resolved npm-launcher environment can reject openhands/claude-* conversations because LiteLLM sends prompt_cache_key. Reproduction needs an approved OpenHands proxy key with an Anthropic model and the affected fresh transitive dependency resolution; a keyless initialization cannot prove provider acceptance. (`F10.default-used`).+- Known issue [#17391](https://github.com/OpenHands/OpenHands/issues/17391): ChatGPT Business subscription login can be refused or use a fallback model with a 404. Completing the entitlement and selected-model checks needs an approved ChatGPT Business workspace login and the reported Docker subscription runtime. (`F10.subscription`).+- Known issue [OpenHands/software-agent-sdk#5181](https://github.com/OpenHands/software-agent-sdk/issues/5181): The SDK can turn an advertised output limit into a requested default cap and silently drop an explicitly enabled reasoning_effort. The narrowed maintainer scope needs an authenticated OpenRouter/proxy route exposing lower-limit endpoints and capability metadata, or a real endpoint accepting the explicit override; a synthetic empty-metadata response would not establish production behavior. (`F10.schema-validation`).+- Known issue [OpenHands/software-agent-sdk#4934](https://github.com/OpenHands/software-agent-sdk/issues/4934): Switching to a MiniMax M3 profile can fail because the SDK sends unsupported reasoning_effort. The runtime switch needs approved working credentials for the starting model and MiniMax route; the reported provider rejection cannot be proved with a loopback connection error. (`F10.schema-validation`, `F10.default-used`).+- Known issue [OpenHands/software-agent-sdk#4613](https://github.com/OpenHands/software-agent-sdk/issues/4613): A namespaced model such as openai/openai/example-model can have its inner provider prefix stripped again in Responses API and reach the provider as example-model. Reproduction needs an approved OpenAI-compatible endpoint serving the namespaced model and its exact stored API-mode settings; a loopback connection refusal cannot prove the outbound identifier. (`F10.custom-model`).+- Known issue [OpenHands/software-agent-sdk#4511](https://github.com/OpenHands/software-agent-sdk/issues/4511): Disabling prompt caching can still emit cache_control on preexisting cache-marked tool messages. The report only supplies a Python serializer reproduction; a Canvas check needs the exact persisted/tool-message precondition and a provider request that exposes the cache markers, with approved model access. (`F10.schema-validation`).+- Known issue [OpenHands/software-agent-sdk#4250](https://github.com/OpenHands/software-agent-sdk/issues/4250): A Cloud Workers AI configuration can fail on context-window validation or be routed as OpenAI with an invalid Cloudflare key. Reproduction needs a disposable OpenHands Cloud account, its Workers AI endpoint/key and the exact saved model configuration; the report supplies no complete configuration steps. (`F10.custom-model`, `F10.default-used`).+- Known issue [OpenHands/software-agent-sdk#4247](https://github.com/OpenHands/software-agent-sdk/issues/4247): An LM Studio configuration can fail with LLM Provider NOT provided although its models endpoint works. Reproduction needs the reachable LM Studio service and the exact model ID/provider-qualified config, which the report does not supply as text; the reported CLI/Docker SDK 1.16.1 installation is not this source stack. (`F10.custom-model`).+- Known issue [OpenHands/software-agent-sdk#5890](https://github.com/OpenHands/software-agent-sdk/issues/5890): Switching the conversation LLM can lose the managed-key refresh-on401 hook. Blocked: needs an approved managed-key refresh source and an allow-listed real model endpoint that returns401 and accepts the refreshed key. Refused loopback cannot establish the refresh/retry outcome; do not supply a fake successful completion. (`F10.edit`, `F10.default-used`).+- Known issue [OpenHands/software-agent-sdk#5862](https://github.com/OpenHands/software-agent-sdk/issues/5862): DeepSeek invalid-key401 responses can be classified as bad-request errors. Blocked by this pass's loopback-only provider routing: a real invalid-key validation request to DeepSeek must be authorized to capture its actual401 shape. No usable key or successful inference is needed, but a loopback connection error does not prove this classification. (`F10.create-validation`).+- Known issue [OpenHands/software-agent-sdk#5838](https://github.com/OpenHands/software-agent-sdk/issues/5838): A live LLM switch during a running tool step can be acknowledged before the new LLM is persisted for GET/restart. Blocked: needs a genuine model-issued bounded tool step (for example `sleep 20`) and an approved starting model. Refused-loopback initialization never enters that step. (`F10.default-used`).+- Known issue [#17806](https://github.com/OpenHands/OpenHands/issues/17806): A model-only Save must preserve an untouched key. Current SDK1.55 refused-loopback validation rejects Save before persistence; the stored key marker remains set. Completing Basic/Advanced preserved-key and runtime-switch proof needs an approved route that accepts preflight. Keep dummy traffic on refused loopback until an external route is authorized; a rejected Save does not establish either key loss or successful preservation through Save. (`F10.edit`).diff --git a/.agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md b/.agents/skills/verify-openhands/references/feature-map/F11-provider-connections.mdindex 8238364b1e..f93d9bbccb 100644--- a/.agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md+++ b/.agents/skills/verify-openhands/references/feature-map/F11-provider-connections.md@@ -60,7 +60,7 @@ Preconditions:- **Cancel (`F11.create-cancel`).** Run `control-openhands browser click 'testid=add-provider-connection'`, `control-openhands browser click 'role=dialog >> role=button[name="Cancel"]'`; `browser count 'testid=provider-connection-modal'` is `0`. Reopen, `control-openhands browser fill 'testid=provider-connection-name-input' QA_escape`, `control-openhands browser press Escape` (count `0`); reopen: `control-openhands browser value 'testid=provider-connection-name-input'` is empty; `control-openhands browser mouse-click 50 500` closes it too. `api GET /api/llm/provider-connections` lists no `QA_escape`.- **Server error (`F11.create-error`).** Run `control-openhands browser click 'testid=add-provider-connection'`, `control-openhands browser fill 'testid=provider-connection-name-input' "QA_$(printf 'x%.0s' $(seq 1 127))"` (130 characters), pick deepseek as above, `control-openhands browser fill 'testid=provider-connection-api-key-input' dummy-key-3`, `control-openhands browser click 'testid=provider-connection-submit'`, then `control-openhands browser toasts` and `control-openhands browser count 'testid=provider-connection-modal'`. The modal stays open (`1`) and the toast reads `String should have at most 128 characters` (pass on main@3500d9e with Agent Server 1.53.0; screenshot `--feature F11.create-error --name toast`). Close with `role=dialog >> role=button[name="Cancel"]`.- **Second connection.** Create `QA_spare` the same way with `dummy-key-2` as key and `control-openhands browser fill 'testid=provider-connection-base-url-input' 'not a url'` (accepted: base URLs are not validated). Toast `Connection "QA_spare" created`.-- **Row menu (`F11.row-menu`, `F11.row-menu-focus`).** Run `control-openhands browser click 'testid=provider-connection-row >> has-text=QA_spare >> testid=provider-connection-menu-trigger'` and `control-openhands browser snapshot 'testid=provider-connection-actions-menu'`: menuitems `Bulk add`, `Edit`, `Delete` (screenshot `--feature F11.row-menu --name open`). `control-openhands browser eval "document.activeElement.getAttribute('data-testid')"` should be `provider-connection-add-models`; on main@3500d9e it stays `provider-connection-menu-trigger` and ArrowDown does nothing until `control-openhands browser press Tab` moves focus to Bulk add ([issue #18204](https://github.com/OpenHands/OpenHands/issues/18204); [draft fix PR #18205](https://github.com/OpenHands/OpenHands/pull/18205)). From there `browser press ArrowDown` focuses `provider-connection-edit`, two `ArrowUp` wrap to `provider-connection-delete`, and `browser press Tab` closes the menu (`browser count 'testid=provider-connection-actions-menu'` `0`). Reopen and `browser press Escape` (`0`); reopen and click the trigger again (`0`); reopen and `control-openhands browser click 'text=Share one API key across multiple models.'` (`0`).+- **Row menu (`F11.row-menu`, `F11.row-menu-focus`).** Run `control-openhands browser click 'testid=provider-connection-row >> has-text=QA_spare >> testid=provider-connection-menu-trigger'` and `control-openhands browser snapshot 'testid=provider-connection-actions-menu'`: menuitems `Bulk add`, `Edit`, `Delete` (screenshot `--feature F11.row-menu --name open`). `control-openhands browser eval "document.activeElement.getAttribute('data-testid')"` should be `provider-connection-add-models`; Known failure (reproduced twice 2026-10-16, [#18204](https://github.com/OpenHands/OpenHands/issues/18204)): it stays `provider-connection-menu-trigger` and ArrowDown does nothing until `control-openhands browser press Tab` moves focus to Bulk add. From there `browser press ArrowDown` focuses `provider-connection-edit`, two `ArrowUp` wrap to `provider-connection-delete`, and `browser press Tab` closes the menu (`browser count 'testid=provider-connection-actions-menu'` `0`). Reopen and `browser press Escape` (`0`); reopen and click the trigger again (`0`); reopen and `control-openhands browser click 'text=Share one API key across multiple models.'` (`0`).- **Edit (`F11.edit`).** Run the trigger click above, then `control-openhands browser click 'testid=provider-connection-actions-menu >> testid=provider-connection-edit'` and `control-openhands browser snapshot 'role=dialog'`: `Edit provider connection`, Name `QA_spare`, Provider `deepseek`, `API Key Leave blank to keep the current key.` with placeholder `<hidden>`, Base URL `not a url`; `browser value 'testid=provider-connection-api-key-input'` is empty and Save is enabled. Run `control-openhands browser fill 'testid=provider-connection-name-input' QA_edited`, `control-openhands browser fill 'testid=provider-connection-base-url-input' 'https://api.deepseek.com'`, `control-openhands browser click 'testid=provider-connection-submit'` and `control-openhands browser wait-text 'Connection "QA_edited" updated'`. After `browser reload` the row reads `QA_edited\ndeepseek\n0 model(s)`, the `QA_spare` count is `0`, and `api GET /api/llm/provider-connections` shows `base_url` `https://api.deepseek.com` and `api_key_set` `true`. Edit again: fill the name with `' '` (`browser enabled 'testid=provider-connection-submit'` `false`), restore `QA_edited`, `control-openhands browser fill 'testid=provider-connection-base-url-input' ''`, Save, `browser wait 'testid=provider-connection-modal' --state detached`; the API shows `base_url` `null`.- **Bulk add (`F11.add-models`).** Run `control-openhands browser click 'testid=provider-connection-row >> has-text=QA_conn >> testid=provider-connection-menu-trigger'` and `control-openhands browser click 'testid=provider-connection-actions-menu >> testid=provider-connection-add-models'`. The dialog `Add models as profiles` has provider `deepseek` selected, `control-openhands browser eval "[...document.querySelectorAll('[data-testid=add-models-connection] option')].map(o=>o.textContent+'='+o.selected).join(' | ')"` is `No connection (keyless)=false | QA_conn=true | QA_edited=false`, and `browser text 'testid=add-models-submit'` is `Add 12 profiles` (every model checked). Run `control-openhands browser uncheck 'testid=add-models-select-all'` (`Add 0 profiles`), `control-openhands browser check 'testid=add-models-check-deepseek/deepseek-flash'` (`Add 1 profiles`), `control-openhands browser click 'testid=add-models-submit'` and `control-openhands browser wait-text 'Added 1'`. After `browser reload` the `QA_conn` row reads `QA_conn\ndeepseek\n1 model(s)` (`QA_edited` stays `0 model(s)`), and `api GET /api/profiles` shows `deepseek-flash` with `provider_connection_id` and `"active_profile": "deepseek-flash"`.- **The agent uses the connection's key (`F11.agent-uses-connection`).** Open Edit on `QA_conn` (as above), leave the key blank, click `testid=provider-connection-submit` and `browser wait-text 'Connection "QA_conn" updated'`. Run `control-openhands conversation start --prompt "Reply with exactly the word: pong" --wait --timeout 180` (prints `<id>`, `"model": "deepseek/deepseek-flash"`, status `finished`), then `control-openhands conversation events <id> --kinds ConversationErrorEvent,MessageEvent`: the agent message is `pong` and there is no `ConversationErrorEvent`.@@ -76,6 +76,12 @@ Preconditions:- **Load error (`F11.load-error`).** With the baseline restored (connections `[]`), run the shell arrange step (it writes only this run's state; the file exists only once the run has created a connection, as the bullets above do, so on a run that never created one `cp` fails: create and delete one through **Add connection** first) `F=$OH_VERIFY_RUN/private/provider-connections/provider_connections.json; cp $F $F.bak; printf '{not json' > $F`. `control-openhands api GET /api/llm/provider-connections` is now `400` (`Provider connections file is unreadable: ...`) while `api GET /api/profiles` is `200`. Run `control-openhands browser reload`, `control-openhands browser text 'testid=provider-connections-load-error'` (`Failed to load provider connections.`), `browser count 'testid=add-provider-connection'` (`1`) and `browser count 'testid=profile-row'` (> `0`; screenshot `--feature F11.load-error --name error`). The message uses the theme's error colour (`browser eval "getComputedStyle(document.querySelector('[data-testid=provider-connections-load-error] p')).color"`: `rgb(231, 106, 94)` on OpenHands-Neutral, `rgb(192, 40, 45)` on Light+, #18330). The profile rows render only because no profile links to a connection at this point: with a linked profile `GET /api/profiles` fails with the same `400` and the list above shows `Failed to load profiles` instead (`F10.list-states`). Restore with `cp $F.bak $F && rm $F.bak`, `browser reload`: `testid=provider-connections-empty` is back and `provider-connections-load-error` count is `0`.- **Errors sweep.** `control-openhands browser errors --app-only` shows `pageErrors` `0`; the HTTP errors are the deliberate `POST 422` (long name), two `DELETE 409` (referenced, then the stale reference) and the `GET 400` from the load-error arrange, all on `/api/llm/provider-connections`. Run `browser errors --clear` afterwards.+- **Redacted key is rejected (`F11.create-validation`, `F11.edit`).** Add `QA_SENTINEL`, choose OpenAI, fill the API Key with the literal `**********` and Base URL with `http://127.0.0.1:1/v1`. Use the native keyboard Save (`browser focus 'testid=provider-connection-submit'`, `browser press Space`; the provider listbox can intercept pointer clicks). Read `api GET /api/llm/provider-connections --pick 0.api_key_set`. Expected: reject the placeholder. Known failure, reproduced 2026-10-12 ([OpenHands/software-agent-sdk#5624](https://github.com/OpenHands/software-agent-sdk/issues/5624)): creation succeeds with `api_key_set: false`. Editing that row, typing the same placeholder and keyboard-saving succeeds and changes it to `true`. A second create under `QA_SENTINEL_2` again succeeds with the key unset. The UI already prevents a whitespace-only key; this recipe drives the sentinel branch, not provider authentication. Remove both unreferenced fixture connections afterwards.++- **Store lock keeps unrelated routes responsive (`F11.list`).** Hold only the own `private/provider-connections/.provider-connections.lock` from an external run-owned process using `fcntl.flock(..., LOCK_EX)` for 34 seconds. Start an authenticated provider-connections list request, then request the own Agent Server's `/alive` two seconds later; reload the actual LLM page and capture its stalled loading state. Expected: `/alive` answers within two seconds while the store request waits. Known failure, reproduced twice 2026-10-12 ([OpenHands/software-agent-sdk#5632](https://github.com/OpenHands/software-agent-sdk/issues/5632)): `/alive` returned 200 after 25.416 and 28.060 seconds. A direct request with a 45-second client deadline received the store's 503 after 30.052 seconds; the CLI's shorter request instead timed out. Release/close the own lock and run `doctor`; do not hold or alter any other run's stores.++- **Connection URL edit reaches next chat (`F11.edit`, `F11.agent-uses-connection`).** Arrange a disposable OpenAI connection `QA_URL_PATCH` with a dummy key and `http://127.0.0.1:1/v1`, plus an activated linked `QA_URL_LINKED` profile without inline key/Base URL. From `/settings/llm`, open the exact connection row menu > `provider-connection-edit`, change `provider-connection-base-url-input` to `http://127.0.0.1:1/v2` and submit; `api GET /api/llm/provider-connections` confirms the stored URL. Start a fresh Home chat with a bounded no-tools prompt, then read only `agent.llm.base_url` in its run-owned `base_state.json`. Repeat the real edit to `/v3` and start another chat. Expected: each new runtime uses the newly saved URL. Known failure, reproduced twice 2026-10-12 ([#17803](https://github.com/OpenHands/OpenHands/issues/17803)): the connection persisted `/v2` and `/v3`, but settings and both new runtimes retained the activated `/v1` snapshot. All model endpoints refuse on loopback; no provider success is needed to observe resolution. Restore the neutral active profile, then delete only the own linked profile and connection.+## Gotchas- Select the Provider by clicking an option (`testid=provider-item-<name>` exists only for verified providers; use `'role=option[name="<label>"]'` for the others), or with `browser choose 'testid=provider-connection-provider-input' <label>`. `browser fill` alone types a filter that is cleared on blur and leaves Save disabled.@@ -84,10 +90,11 @@ Preconditions:- Edit's Save is enabled even with no changes, and a blank key on Edit is omitted from the request (the stored key is kept); there is no way to clear a key.- The delete spinner (`aria-busy`) lasts under 100 ms on a local stack; `--observe` sees only `Delete` then `<absent>`.- `role=dialog` matches two nested elements in these modals; count `> 0`, do not expect `1`.-- The first profile added on a run with no profiles becomes the active one. Activating a linked profile copies the resolved key into the active agent settings, which keep the connection id: that is why the server still counts a reference after every linked profile is gone (see `F11.delete-stale-reference`) (OpenHands/software-agent-sdk#5498).+- The first profile added on a run with no profiles becomes the active one. Activating a linked profile copies the resolved key into the active agent settings, which keep the connection id: that is why the server still counts a reference after every linked profile is gone (see `F11.delete-stale-reference`) ([OpenHands/software-agent-sdk#5498](https://github.com/OpenHands/software-agent-sdk/issues/5498)).- Delete a connection's linked profiles before the connection; then also make an unlinked profile active, or the delete is refused.- Rotation results are read from the conversation (`conversation events ... --kinds ConversationErrorEvent,MessageEvent` and the banner in the chat), not from the settings page, which never shows the key.- Model Router's **Add connection** reuses this modal (F12); its toasts and fields are the same.- `F11.load-error` needs a real server failure: the recipe corrupts this run's own `provider_connections.json` and restores it. Always restore it (keep the `.bak`), or every later family on the run sees the error; never touch another run's state.- At phone width a long name squeezes the provider column to one letter (`d…`); the full values stay in the API and the name's `title`.-- Known issue OpenHands/software-agent-sdk#5497: after a key rotation, new conversations still run agent steps on the old key; only title generation uses the new one (`F11.rotate`).+- Known issue [OpenHands/software-agent-sdk#5497](https://github.com/OpenHands/software-agent-sdk/issues/5497): after a key rotation, new conversations still run agent steps on the old key; only title generation uses the new one (`F11.rotate`).+- Known issue [OpenHands/software-agent-sdk#5207](https://github.com/OpenHands/software-agent-sdk/issues/5207): OpenRouter linked profiles can lose connection credentials at raw agent/condenser startup or reuse stale credentials after rotation; opaque reasoning can also be lost on persisted turns. End-to-end checks need an approved OpenRouter connection and a model returning reasoning_details, including the linked-profile switch and reload path; loopback failure does not prove them. (`F11.agent-uses-connection`, `F11.rotate`).diff --git a/.agents/skills/verify-openhands/references/feature-map/F12-model-router.md b/.agents/skills/verify-openhands/references/feature-map/F12-model-router.mdindex 43e977b2a3..3a8e320d25 100644--- a/.agents/skills/verify-openhands/references/feature-map/F12-model-router.md+++ b/.agents/skills/verify-openhands/references/feature-map/F12-model-router.md@@ -82,6 +82,40 @@ Preconditions:- **Restart persistence (`F12.list`).** Run `control-openhands restart`, `control-openhands doctor`, `control-openhands browser goto /settings/meta-llm` and `control-openhands browser count 'testid=meta-profile-row-QA_router_phone >> testid=meta-profile-active-badge'`; it is `1`.- **Cleanup.** Switch **Run on first message** off through its label first (it disables once no router is left), then delete each `QA_*` router through the row menu. Remove the arranged LLM profile and connection with `control-openhands api DELETE /api/profiles/deepseek-chat --write` and `control-openhands api DELETE /api/llm/provider-connections/<connection id> --write` (`<connection id>` from `control-openhands api GET /api/llm/provider-connections`), or through Settings → LLM (F10).+- **Router activation is honest on ACP (`F12.activate`).** ACP cannot attach the router; activation must report that refusal rather than plain success. Known failure (reproduced twice 2026-10-16, [OpenHands/software-agent-sdk#5879](https://github.com/OpenHands/software-agent-sdk/issues/5879)): actual global ACP settings produced an activated toast while both stored pointers stayed null. An active ACP profile pointer alone does not create this condition. The block restores the original owned settings bytes; its assertion rejects the misleading success toast.++ ```sh known-bug F12.activate acp-settings+ QA_ACP_SETTINGS="$OH_VERIFY_RUN/private/qa-known-acp-settings.json"+ cp "$OH_VERIFY_RUN/private/settings.json" "$QA_ACP_SETTINGS"+ trap 'cp "$QA_ACP_SETTINGS" "$OH_VERIFY_RUN/private/settings.json"; rm -f "$QA_ACP_SETTINGS"; control-openhands api DELETE /api/meta-profiles/QA_KNOWN_ACP_ROUTER --write >/dev/null; control-openhands browser reload >/dev/null' EXIT+ control-openhands api POST /api/meta-profiles/QA_KNOWN_ACP_ROUTER --data '{"classifier_model":"QA_KNOWN_CLASSIFIER","prompt_template":"Pick {{ instance_text }}"}' --write+ control-openhands api PATCH /api/settings --data '{"agent_settings_diff":{"agent_kind":"acp"}}' --write --pick agent_settings.agent_kind+ control-openhands assert --field value --equals acp -- api GET /api/settings --pick agent_settings.agent_kind+ control-openhands browser viewport desktop+ control-openhands browser goto /settings/meta-llm+ control-openhands browser wait 'testid=meta-profile-row-QA_KNOWN_ACP_ROUTER'+ control-openhands browser click 'testid=meta-profile-menu-trigger-QA_KNOWN_ACP_ROUTER'+ control-openhands browser click 'testid=meta-profile-set-active' --observe 'role=status' --observe-ms 1000+ control-openhands assert --field value --equals false --bug-shows true -- browser eval 'document.body.innerText.includes("Meta-profile \"QA_KNOWN_ACP_ROUTER\" activated")' # bug+ ```++- **Unreadable settings refuse router activation (`F12.activate`).** Expected: a handled409 without a filesystem path and no store overwrite. Known failure (reproduced twice 2026-10-16, [OpenHands/software-agent-sdk#5922](https://github.com/OpenHands/software-agent-sdk/issues/5922)): the real Set active path showed the generic500 and the settings path. The corrupt bytes survived. Back up only this run's settings, and restore them in `EXIT` before the runner's health check.++ ```sh known-bug F12.activate corrupt-settings+ QA_ROUTER_SETTINGS="$OH_VERIFY_RUN/private/qa-known-router-settings.json"+ cp "$OH_VERIFY_RUN/private/settings.json" "$QA_ROUTER_SETTINGS"+ trap 'cp "$QA_ROUTER_SETTINGS" "$OH_VERIFY_RUN/private/settings.json"; rm -f "$QA_ROUTER_SETTINGS"; control-openhands api DELETE /api/meta-profiles/QA_KNOWN_CORRUPT_ROUTER --write >/dev/null; control-openhands browser reload >/dev/null' EXIT+ control-openhands api POST /api/meta-profiles/QA_KNOWN_CORRUPT_ROUTER --data '{"classifier_model":"QA_KNOWN_CLASSIFIER","prompt_template":"Pick {{ instance_text }}"}' --write+ control-openhands browser viewport desktop+ control-openhands browser goto /settings/meta-llm+ control-openhands browser wait 'testid=meta-profile-row-QA_KNOWN_CORRUPT_ROUTER'+ printf '{"agent_settings": ' > "$OH_VERIFY_RUN/private/settings.json"+ control-openhands browser click 'testid=meta-profile-menu-trigger-QA_KNOWN_CORRUPT_ROUTER'+ control-openhands browser network --clear+ control-openhands browser click 'testid=meta-profile-set-active' --observe 'role=status' --observe-ms 1000+ control-openhands assert --field recent.0.status --equals 409 --bug 'recent.0.status eq 500 && recent.0.path eq "/api/meta-profiles/QA_KNOWN_CORRUPT_ROUTER/activate"' -- browser network --filter '/api/meta-profiles/QA_KNOWN_CORRUPT_ROUTER/activate' # bug+ ```+## Gotchas- Settings nav test ids embed the route: quote `'testid=sidebar-settings-/settings/meta-llm'` as written.@@ -96,4 +130,5 @@ Preconditions:- `service stop agent-server` does not show this page's error state: the whole app switches to the Manage backends card (F02). Run `control-openhands restart` afterwards.- The selection in the classifier and connection pickers is a combobox: open it with a click on the input, then click `role=option[name="..."]`; `browser fill` on the classifier sets a free-text value (allowed) instead of picking a profile.- At 390 px the row truncates the name (`QA_route…`) and summary; read names with `browser text`, not the screenshot.-- The Model table placeholder showed a literal `\n` on one line until #18035 (closed #17939); a checkout without that fix still does (`F12.template-custom`).+- The Model table placeholder displays separate lines (`F12.template-custom`).+- Known issue [OpenHands/software-agent-sdk#5878](https://github.com/OpenHands/software-agent-sdk/issues/5878): Router files can be written under a literal `~` directory when Agent Server was configured with leading-tilde `OH_PERSISTENCE_DIR`. Blocked on the normal absolute-path launcher fixture: needs a disposable server with private HOME/cwd and literal `~/...` or relative persistence configuration, then a router save and filesystem readback. Never point it at user state. (`F12.create-first`, `F12.list`).diff --git a/.agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md b/.agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.mdindex ff1fff2f5a..e49e45d15a 100644--- a/.agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md+++ b/.agents/skills/verify-openhands/references/feature-map/F13-agent-profiles.md@@ -20,7 +20,7 @@ Source: `src/routes/agent-profiles-settings.tsx`, `src/routes/agent-settings.tsx- `F13.stale-llm-ref`: editing a profile whose LLM profile no longer exists preselects the active LLM profile and enables Save, so saving heals the reference.- `F13.set-active`: **Set as active** toasts `Switched to profile "<name>"`, moves the Default badge and is disabled on the active row.- `F13.active-drives-conversation`: a conversation started from Home after activation launches from that profile (its id and settings are stamped on the conversation).-- `F13.llm-pill-precedence`: when the active profile pins a different LLM profile than the Home LLM pill, the launch uses the pill's model and not the profile's other settings (by design).+- `F13.llm-pill-precedence`: when the active profile pins a different LLM profile than the Home LLM pill, a backend advertising per-launch LLM override uses the pill's LLM while preserving the named profile's other settings and identity.- `F13.mcp-scope`: **MCP servers** offers **All servers** (every configured server listed, switched on and disabled) or **Choose servers** (a switch per configured server); the choice persists; a chosen server that was removed stays listed with a red warning, and launching with it is refused.- `F13.secrets-scope`: **Secrets** offers **No profile restriction** (every switch on and disabled) or **Choose secrets**; on an ACP profile, Choose preselects that provider's secret names; the selection persists.- `F13.acp-form`: choosing **ACP (external subprocess)** shows Preset, Command, Credentials and Model; each preset prefills its command and its model choice (`Agent default`, or Gemini CLI's preferred `gemini-2.5-pro`), **Custom** empties the command and disables Save until one is typed, typing a preset's exact command selects that preset, and the Model list's **Custom** opens a free-text model field.@@ -29,7 +29,7 @@ Source: `src/routes/agent-profiles-settings.tsx`, `src/routes/agent-settings.tsx- `F13.acp-credentials`: built-in presets show a Credentials section (password fields), an auth banner (signed in / configured), a conflict warning for mutually exclusive Claude credentials, and `Already saved — leave blank to keep` for stored values. Credentials are saved as global secrets.- `F13.acp-preset-credentials`: the Credentials fields follow the preset: Claude Code `CLAUDE_CODE_OAUTH_TOKEN`, `ANTHROPIC_API_KEY`, `ANTHROPIC_BASE_URL`; Codex `OPENAI_API_KEY`, `CODEX_AUTH_JSON` (multi-line), `OPENAI_BASE_URL`; Gemini CLI `GOOGLE_APPLICATION_CREDENTIALS_JSON` (multi-line), `GOOGLE_CLOUD_PROJECT`, `GOOGLE_CLOUD_LOCATION`, `GOOGLE_GENAI_USE_VERTEXAI`, `GEMINI_API_KEY`, `GEMINI_BASE_URL`; Pi `PI_AUTH_JSON` (multi-line), `ANTHROPIC_API_KEY`; OpenCode `OPENCODE_AUTH_CONTENT` (multi-line), `OPENCODE_API_KEY`; **Custom** has no Credentials section. With no host login and no stored credential (Gemini here) no auth banner shows.- `F13.menu-keyboard`: opening the row's `...` menu focuses Edit; ArrowDown/ArrowUp move focus between the enabled items (wrapping, and skipping the disabled **Set as active** on the active row), Enter runs the focused item, and Tab, Escape or a click outside close the menu.-- `F13.delete`: Delete asks for confirmation; Cancel keeps the row, Delete removes it. Deleting the active profile leaves none active; deleting the last profile makes the server re-seed `default`.+- `F13.delete`: Delete asks for confirmation; Cancel keeps the row, Delete removes it. Deleting the active profile selects `default` immediately; deleting the last profile makes the server re-seed and activate `default`.- `F13.phone`: the list, row menu and editor fit a 390 px viewport without horizontal overflow.- `F13.acp-conversation`: a conversation launched from an active ACP profile runs the external agent. Blocked without an approved provider account.- `F13.cloud-read-only`: on a Cloud backend, organization members see rows without the `...` menu or **Add agent profile**. Blocked without a Cloud account.@@ -54,45 +54,78 @@ Preconditions:- Dropdowns are autocompletes. Pick an option with `control-openhands browser choose 'testid=<testid>' '<label>'` (proven on `agent-type-selector`, `agent-preset-selector`, `agent-model-selector` (`Custom`), `agent-profile-llm-selector`, `agent-settings-mcp-mode` and `agent-settings-secrets-mode`). The manual opener, `control-openhands browser click 'div:has(> div > input[data-testid=<testid>]) >> role=button[name="Show suggestions"]'` then `browser click 'role=option[name="<label>"]'`, works for most of them but is flaky right after an editor opens (see Gotchas); clicking the input itself can open an empty, filtered list.- **List (`F13.list`).** From `/` run `control-openhands browser click 'testid=backend-selector-settings-link' --expect-url '/settings'`, then `control-openhands browser url` (`/settings/agents`) and `control-openhands browser text 'testid=agent-profile-row'`: `default`, its LLM profile `default`, `Default` on three lines. `control-openhands browser count 'testid=agent-profile-active-badge'` is `1`; `control-openhands browser screenshot --feature F13.list --name list`. Command menu: run `control-openhands browser goto /` (the browser is on `/settings/agents`, where the URL wait would pass at once), `control-openhands browser press Control+k`, `control-openhands browser type 'testid=command-menu >> role=combobox' 'Agent settings'`, `control-openhands browser press Enter`, `control-openhands browser wait-url '/settings/agents(\?|$)'`. Legacy URL: `control-openhands browser goto /settings/agent` returns `.../settings/agents`.-- **No LLM profile (`F13.editor-no-llm`).** Before the preset, run `control-openhands browser click 'testid=add-agent-profile'`, then `control-openhands browser text 'testid=agent-profile-no-llm'`: `Create an LLM profile first to define an OpenHands agent.` `control-openhands browser text 'testid=agent-profile-editor-title'` is `Add agent profile` and `control-openhands browser text 'testid=agent-profile-editor-description'` is `Configure the agent below, then click Save to create this profile.` (agent copy, not the LLM page's hint, since #18035). Run `control-openhands browser fill 'testid=agent-profile-name-input' QA_NoLlm`; Save turns enabled (`control-openhands browser enabled 'testid=save-agent-profile-btn'`). Click `testid=save-agent-profile-btn` and run `control-openhands browser toasts`: `Select an LLM profile for this agent.` `api GET /api/agent-profiles` lists no `QA_NoLlm`. Leave with `control-openhands browser click 'testid=cancel-agent-profile-btn'`.+- **No LLM profile (`F13.editor-no-llm`).** Before the preset, run `control-openhands browser click 'testid=add-agent-profile'`, then `control-openhands browser text 'testid=agent-profile-no-llm'`: `Create an LLM profile first to define an OpenHands agent.` `control-openhands browser text 'testid=agent-profile-editor-title'` is `Add agent profile` and `control-openhands browser text 'testid=agent-profile-editor-description'` is `Configure the agent below, then click Save to create this profile.`. Run `control-openhands browser fill 'testid=agent-profile-name-input' QA_NoLlm`; Save turns enabled (`control-openhands browser enabled 'testid=save-agent-profile-btn'`). Click `testid=save-agent-profile-btn` and run `control-openhands browser toasts`: `Select an LLM profile for this agent.` `api GET /api/agent-profiles` lists no `QA_NoLlm`. Leave with `control-openhands browser click 'testid=cancel-agent-profile-btn'`.- **Name rules (`F13.name-validation`).** In the Add editor, fill `testid=agent-profile-name-input` with `''`, `default`, `'bad name!'` and `QA_ok` in turn; after each run `control-openhands browser attr 'testid=agent-profile-name-input' aria-invalid` and `control-openhands browser enabled 'testid=save-agent-profile-btn'`. Results: `''` → `true`/`false`; `default` (duplicate) → `true`/`false`; `'bad name!'` → `true`/`false`; `QA_ok` → `false`/`true`. The rule under the field is always shown (grey) and turns the theme danger colour (`text-danger`, #18199) while the name is invalid; it reads `Profile name must start with a letter or digit, and contain only letters, digits, dots, underscores, or hyphens (max 64 characters)`. For the duplicate, `control-openhands browser count 'role=alert >> text=already exists'` is `1`: the alert is the field's description, in the danger colour (`rgb(231, 106, 94)` on OpenHands-Neutral, `rgb(209, 52, 56)` on Light+), while the rule text stays grey. It reads `An agent profile with this name already exists. Choose a different name, or edit the existing one.` Take `control-openhands browser screenshot --feature F13.name-validation --name duplicate` with `default` filled. In Edit the same message appears for another profile's name and not for the profile's own name (`F13.edit`).- **ACP form (`F13.acp-form`).** In the Add editor run `control-openhands browser click 'testid=agent-type-selector'` and `control-openhands browser click 'role=option[name="ACP (external subprocess)"]'`. `control-openhands browser value 'testid=agent-preset-selector'` is `Claude Code`, `browser value 'testid=agent-command-input'` is `npx -y --prefer-offline @agentclientprotocol/[email protected]` and `browser value 'testid=agent-model-selector'` is `Agent default`; `agent-profile-no-llm` is gone (ACP needs no LLM profile). The preset list (`control-openhands browser click 'div:has(> div > input[data-testid=agent-preset-selector]) >> role=button[name="Show suggestions"]'`, then `control-openhands browser snapshot 'role=listbox'`) is `Claude Code`, `Codex`, `Gemini CLI`, `Pi`, `OpenCode`, `Custom` (`browser press Escape`). Pick `Codex` and `Gemini CLI` from `testid=agent-preset-selector`: the command becomes `npx -y --prefer-offline @agentclientprotocol/[email protected]` / `npx -y --prefer-offline @google/[email protected] --acp` and the model `Agent default` / `gemini-2.5-pro` (the models come from the agent since #18131: `F13.acp-model-discovery`). Pick `Custom`: the command is empty and Save is disabled. `control-openhands browser fill 'testid=agent-command-input' 'npx -y qa-custom-acp --stdio'` enables Save. Filling the exact Codex command flips the preset to `Codex`; appending ` --verbose` flips it back to `Custom`. Model override: with a built-in preset, run `control-openhands browser choose 'testid=agent-model-selector' 'Custom'` (or open the model list with its toggle button and `control-openhands browser click 'role=option[name="Custom"][exact]'`); `control-openhands browser count 'testid=agent-model-input'` is `1` and empty. The **Custom** preset shows `agent-model-input` directly (no model list).- **Create an ACP profile (`F13.create`).** With preset `Custom`, fill `testid=agent-profile-name-input` with `QA_acp_custom`, `testid=agent-command-input` with `'npx -y qa-custom-acp --stdio'` and `testid=agent-model-input` with `qa-model`, click `testid=save-agent-profile-btn`, then `control-openhands browser wait-text 'Profile "QA_acp_custom" created' --timeout 10000`, `control-openhands browser reload` and `control-openhands browser text 'testid=agent-profile-row >> has-text=QA_acp_custom'`: `QA_acp_custom` / `ACP (external subprocess)`. `control-openhands api GET /api/agent-profiles/QA_acp_custom` shows `acp_server` `custom`, the command and `acp_model` `qa-model`.- **Pi and OpenCode presets (`F13.acp-local-presets`).** For each of `OpenCode` and `Pi`: run `control-openhands browser click 'testid=add-agent-profile'`, `control-openhands browser choose 'testid=agent-type-selector' 'ACP (external subprocess)'` and `control-openhands browser choose 'testid=agent-preset-selector' '<preset>'`. `browser value 'testid=agent-command-input'` is `npx -y --prefer-offline [email protected] acp` / `npx -y --prefer-offline [email protected] --package=@earendil-works/[email protected] pi-acp` (the versions follow `@openhands/typescript-client`). Fill `testid=agent-profile-name-input` with `QA_opencode` / `QA_pi`, run `control-openhands browser wait 'testid=agent-model-discovering' --state detached --timeout 120000`, click `testid=save-agent-profile-btn` and `control-openhands browser wait-text 'Profile "<name>" created' --timeout 10000`. `control-openhands api GET /api/agent-profiles/<name>` has `acp_server` `opencode` / `pi`, `acp_command` `null` and `acp_model` `null`, and after `browser reload` both rows read `<name>` / `ACP (external subprocess)`. Cloud hides both presets (blocked here: needs a Cloud backend).- **Models from the agent (`F13.acp-model-discovery`).** In a new ACP editor, each built-in preset pick first shows (unless a successful lookup for that preset and secret set is cached, about 5 minutes: OpenCode then skips it) `control-openhands browser text 'testid=agent-model-discovering'` `Asking <agent> which models it offers…` (the agent-server runs the agent CLI through npx for `POST /api/acp/models`; the first lookup can take a minute while npx downloads it). Wait with `control-openhands browser wait 'testid=agent-model-discovering' --state detached --timeout 120000`, then read `control-openhands browser testids --hidden --filter agent-model`. In this sandbox (no Claude, Codex, Gemini or Pi login for the agent-server): Claude Code fails to start and shows `agent-model-list-after-first-conversation` (`Claude Code's models appear here after your first conversation with it. You can also switch models from the chat.`); Codex, Gemini CLI and Pi need a login and show `agent-model-discovery-needs-auth` (`Sign in or add credentials above to see the models <agent> offers.`), with the model left at `Agent default` (`gemini-2.5-pro` for Gemini); `<run>/private/stack.log` logs `ACP model discovery failed: server=<key> code=ACPInitError|ACPAuthRequired`. OpenCode needs no key: `browser value 'testid=agent-model-selector'` is `Agent default (OpenCode/Big Pickle)` and the list (open it with `browser click 'div:has(> div > input[data-testid=agent-model-selector]) >> role=button[name="Show suggestions"]'`, then `browser snapshot 'role=listbox'`) has that entry, a non-empty list of the models OpenCode reports live (12 on 2026-10-13, such as `OpenCode/Big Pickle` and `OpenCode/Exo Free`) and `Custom` (screenshot `--feature F13.acp-model-discovery --name opencode-models`). Edit `QA_opencode`: Save is disabled until `control-openhands browser choose 'testid=agent-model-selector' 'OpenCode/Exo Free'`; Save (`Profile "QA_opencode" updated`) stores `acp_model` `opencode/exo-free`, and reopening shows `OpenCode/Exo Free` with Save disabled. With a saved `OPENAI_API_KEY` (even a dummy, `F13.acp-credentials`) Codex lists its models and shows `Agent default (GPT-6-Astra)`. Lookups never add a conversation to the list (`api GET "/api/conversations/search?limit=20"` stays empty).- **Edit and rename (`F13.edit`).** Run `control-openhands browser click 'testid=agent-profile-row >> has-text=QA_acp_custom >> testid=agent-profile-menu-trigger'`; `control-openhands browser snapshot 'testid=agent-profile-actions-menu'` lists `Edit`, `Set as active`, `Delete`. Click `testid=agent-profile-edit`: the title is `Edit agent profile`, the description `Editing profile "QA_acp_custom" - save to apply changes`, the type/preset/command/model values are the stored ones, and Save is disabled. Note the `id` from `api GET /api/agent-profiles`, fill the name with `QA_acp_renamed`, click Save, `control-openhands browser wait-text 'Profile "QA_acp_renamed" updated' --timeout 10000` and `browser reload`: `browser count 'testid=agent-profile-row >> has-text=QA_acp_custom'` is `0`, the renamed row is `1` and `api GET /api/agent-profiles` shows the same id under the new name. Back/Cancel: open any editor, change a field, click `testid=back-to-agent-profiles` (or `testid=cancel-agent-profile-btn`); `browser count 'testid=add-agent-profile'` is `1` and the API is unchanged.-- **Credentials (`F13.acp-credentials`).** In a new ACP editor with preset `Claude Code`, `control-openhands browser testids --hidden --filter acp` lists `settings-acp-secret-CLAUDE_CODE_OAUTH_TOKEN`, `-ANTHROPIC_API_KEY`, `-ANTHROPIC_BASE_URL` and one banner (`settings-acp-auth-detected` when the agent-server machine has a Claude CLI login, as this sandbox does: `You're already signed in to Claude Code — you can leave the fields below blank.`). `control-openhands browser attr 'testid=settings-acp-secret-ANTHROPIC_API_KEY' type` is `password`. Fill `settings-acp-secret-CLAUDE_CODE_OAUTH_TOKEN` with `qa-dummy-oauth` and `settings-acp-secret-ANTHROPIC_API_KEY` with `qa-dummy-key`; `control-openhands browser text 'testid=acp-credential-conflict-warning'` is `Setting ANTHROPIC_API_KEY alongside CLAUDE_CODE_OAUTH_TOKEN breaks its authentication — leave one of them blank.` (screenshot `--feature F13.acp-credentials --name conflict` after `browser scroll 'testid=acp-credential-conflict-warning'`). Known failure in Light+ (reproduced 2026-10-13 on TARGET and BASE, #18004): the auth banners use raw `text-green-200` (`settings-acp-auth-detected`) and `text-blue-200` (`settings-acp-auth-configured`) on a pale tint over white (`acp-auth-status-banner.tsx:63/93`), measuring 1.10:1 and 1.26:1; expected: readable banner text in every theme (screenshot `--feature F13.acp-credentials --name detected-banner-lightplus`). Cancel; `api GET /api/settings/secrets` gains nothing. Saved credential: new profile `QA_codex`, ACP, preset `Codex`, fill `testid=settings-acp-secret-OPENAI_API_KEY` with `qa-dummy-openai`, Save (`Profile "QA_codex" created`); `api GET /api/settings/secrets` now lists `OPENAI_API_KEY` and the profile has `acp_server` `codex`, `acp_command` `null`, `acp_model` `null` (the agent's own default since #18131). Reopen it: `control-openhands browser testids --hidden --filter settings-acp-auth` shows `settings-acp-auth-configured` (`Credentials for Codex are configured — you can leave the fields below blank.`) and `control-openhands browser attr 'testid=settings-acp-secret-OPENAI_API_KEY' placeholder` is `Already saved — leave blank to keep` with an empty value.+- **Credentials (`F13.acp-credentials`).** In a new ACP editor with preset `Claude Code`, `control-openhands browser testids --hidden --filter acp` lists `settings-acp-secret-CLAUDE_CODE_OAUTH_TOKEN`, `-ANTHROPIC_API_KEY`, `-ANTHROPIC_BASE_URL` and one banner (`settings-acp-auth-detected` when the agent-server machine has a Claude CLI login: `You're already signed in to Claude Code — you can leave the fields below blank.`). `control-openhands browser attr 'testid=settings-acp-secret-ANTHROPIC_API_KEY' type` is `password`. Fill `settings-acp-secret-CLAUDE_CODE_OAUTH_TOKEN` with `qa-dummy-oauth` and `settings-acp-secret-ANTHROPIC_API_KEY` with `qa-dummy-key`; `control-openhands browser text 'testid=acp-credential-conflict-warning'` is `Setting ANTHROPIC_API_KEY alongside CLAUDE_CODE_OAUTH_TOKEN breaks its authentication — leave one of them blank.` (screenshot `--feature F13.acp-credentials --name conflict` after `browser scroll 'testid=acp-credential-conflict-warning'`). Known failure in Light+ (reproduced 2026-10-13 on TARGET and BASE, #18004): the auth banners use raw `text-green-200` (`settings-acp-auth-detected`) and `text-blue-200` (`settings-acp-auth-configured`) on a pale tint over white (`acp-auth-status-banner.tsx:63/93`), measuring 1.10:1 and 1.26:1; expected: readable banner text in every theme (screenshot `--feature F13.acp-credentials --name detected-banner-lightplus`). Cancel; `api GET /api/settings/secrets` gains nothing. Saved credential: new profile `QA_codex`, ACP, preset `Codex`, fill `testid=settings-acp-secret-OPENAI_API_KEY` with `qa-dummy-openai`, Save (`Profile "QA_codex" created`); `api GET /api/settings/secrets` now lists `OPENAI_API_KEY` and the profile has `acp_server` `codex`, `acp_command` `null`, `acp_model` `null` (the agent's own default since #18131). Reopen it: `control-openhands browser testids --hidden --filter settings-acp-auth` shows `settings-acp-auth-configured` (`Credentials for Codex are configured — you can leave the fields below blank.`) and `control-openhands browser attr 'testid=settings-acp-secret-OPENAI_API_KEY' placeholder` is `Already saved — leave blank to keep` with an empty value.- **Secrets scope (`F13.secrets-scope`).** In the `QA_codex` editor, `control-openhands browser eval "[...document.querySelectorAll('[data-testid=agent-settings-secret-list] input')].map(i=>i.closest('li').innerText.split('\n')[0]+'='+i.checked+(i.disabled?'(disabled)':''))"` shows every name `=true(disabled)` under `No profile restriction`. Open `testid=agent-settings-secrets-mode`, `control-openhands browser click 'role=option[name="Choose secrets"]'` and rerun the eval: `OPENHANDS_AUTOMATION_API_KEY=false`, `OPENAI_API_KEY=true`, `CODEX_AUTH_JSON=true`, `OPENAI_BASE_URL=true`. Save (`Profile "QA_codex" updated`); `api GET /api/agent-profiles/QA_codex` has those three in `secret_refs`, and after `browser reload` and Edit the mode and switches are the same. On an OpenHands profile, Choose starts with nothing selected; a switch toggles by its name: in the Add editor pick `Choose secrets`, then `control-openhands browser click 'testid=agent-settings-secret-list >> text=OPENHANDS_AUTOMATION_API_KEY'` turns `document.querySelector('[data-testid=agent-settings-secret-OPENHANDS_AUTOMATION_API_KEY]').checked` from `false` to `true`.- **Stale LLM reference (`F13.stale-llm-ref`).** After `control-openhands llm preset deepseek`, run `control-openhands browser reload`, open `testid=agent-profile-row >> has-text=default >> testid=agent-profile-menu-trigger` → `testid=agent-profile-edit`, `control-openhands browser wait 'testid=agent-profile-llm-selector'`. `browser value 'testid=agent-profile-llm-selector'` is `deepseek-flash (deepseek/deepseek-flash)` and Save is enabled although nothing was touched. Save (`Profile "default" updated`), `browser reload`: the row reads `default` / `deepseek-flash` / `Default` and `api GET /api/agent-profiles/default` has `llm_profile_ref` `deepseek-flash` with the same id. Screenshot `--feature F13.stale-llm-ref --name fallback` before saving.-- **OpenHands options (`F13.openhands-options`, `F13.create`).** Click `testid=add-agent-profile`, fill the name `QA_oh`. Defaults: `browser value` on `testid=agent-profile-llm-selector` is the active LLM profile (`deepseek-flash (deepseek/deepseek-flash)`), on `testid=agent-settings-system-prompt-mode` `OpenHands default`, on `testid=agent-settings-tools-mode` `Standard tools` and on `testid=sdk-settings-tool_concurrency_limit` `1`; `control-openhands browser text 'testid=agent-settings-tool-list'` lists `file_editor`, `task_tracker`, `terminal` and `switch_llm` with their descriptions (the list follows the Agent Server; 1.53.0 here). Run `control-openhands browser choose 'testid=agent-settings-tools-mode' 'Choose tools'`; `control-openhands browser eval "[...document.querySelectorAll('[data-testid=agent-settings-tool-list] input')].map(i=>i.dataset.testid.replace('agent-settings-tool-','')+'='+i.checked)"` is `file_editor=true`, `task_tool_set=false`, `task_tracker=true`, `terminal=true`, `glob=false`, `grep=false`, `switch_llm=true`. Click `'testid=agent-settings-tool-list >> text=switch_llm'` and `'testid=agent-settings-tool-list >> text=task_tool_set'` (the eval flips both). Fill `testid=sdk-settings-tool_concurrency_limit` with `0` and click Save: `browser toasts` shows `Parallel tool calls must be at least 1` and `browser eval "document.querySelector('[data-testid=sdk-settings-tool_concurrency_limit]').validationMessage"` is `Value must be greater than or equal to 1.` Fill `2`, `control-openhands browser choose 'testid=agent-profile-llm-selector' 'deepseek-pro (deepseek/deepseek-v4-pro)'`, Save (`Profile "QA_oh" created`), `browser reload`: the row reads `QA_oh` / `deepseek-pro`, and `api GET /api/agent-profiles/QA_oh` has `llm_profile_ref` `deepseek-pro`, `tools` named `file_editor`, `task_tool_set`, `task_tracker`, `terminal`, `enable_classify_and_switch_llm_tool` `false` and `tool_concurrency_limit` `2`. Reopening shows `Choose tools` with the same switches, `2` and `deepseek-pro`, with Save disabled. (There are no **Enable sub-agents** or **Let the agent switch LLM profiles** switches: `browser count 'text=Enable sub-agents'` is `0`.)+- **OpenHands options (`F13.openhands-options`, `F13.create`).** Click `testid=add-agent-profile`, fill the name `QA_oh`. Defaults: `browser value` on `testid=agent-profile-llm-selector` is the active LLM profile (`deepseek-flash (deepseek/deepseek-flash)`), on `testid=agent-settings-system-prompt-mode` `OpenHands default`, on `testid=agent-settings-tools-mode` `Standard tools` and on `testid=sdk-settings-tool_concurrency_limit` `1`; `control-openhands browser text 'testid=agent-settings-tool-list'` lists `file_editor`, `task_tracker`, `terminal` and `switch_llm` with their descriptions (the list follows the Agent Server; it also includes `browser_tool_set` when usable). Run `control-openhands browser choose 'testid=agent-settings-tools-mode' 'Choose tools'`; `control-openhands browser eval "[...document.querySelectorAll('[data-testid=agent-settings-tool-list] input')].map(i=>i.dataset.testid.replace('agent-settings-tool-','')+'='+i.checked)"` is the selected standard tools, including `file_editor=true`, `task_tracker=true`, `terminal=true` and `switch_llm=true`; `task_tool_set=false`, `glob=false` and `grep=false`. Additional rows follow the current server catalog. Click `'testid=agent-settings-tool-list >> text=switch_llm'` and `'testid=agent-settings-tool-list >> text=task_tool_set'` (the eval flips both). Fill `testid=sdk-settings-tool_concurrency_limit` with `0` and click Save: `browser toasts` shows `Parallel tool calls must be at least 1` and `browser eval "document.querySelector('[data-testid=sdk-settings-tool_concurrency_limit]').validationMessage"` is `Value must be greater than or equal to 1.` Fill `2`, `control-openhands browser choose 'testid=agent-profile-llm-selector' 'deepseek-pro (deepseek/deepseek-v4-pro)'`, Save (`Profile "QA_oh" created`), `browser reload`: the row reads `QA_oh` / `deepseek-pro`, and `api GET /api/agent-profiles/QA_oh` has `llm_profile_ref` `deepseek-pro`, `tools` retaining the selected standard tools plus `task_tool_set` (including `task_tracker`), `enable_classify_and_switch_llm_tool` `false` and `tool_concurrency_limit` `2`. Reopening shows `Choose tools` with the same switches, `2` and `deepseek-pro`, with Save disabled. (There are no **Enable sub-agents** or **Let the agent switch LLM profiles** switches: `browser count 'text=Enable sub-agents'` is `0`.)System prompt: reopen `QA_oh`, `control-openhands browser choose 'testid=agent-settings-system-prompt-mode' 'Default + your instructions'` and fill `testid=agent-settings-system-prompt` with `Use terse QA answers.`. Save, `browser reload` and reopen: `api GET /api/agent-profiles/QA_oh` has `system_message_suffix` `Use terse QA answers.` and `persona` `null`, and the editor keeps the mode and text with Save disabled. Then choose `Custom persona`, fill `You are the disposable QA persona.`, Save, reload and reopen: `persona` is that text, `system_message_suffix` is `null`, and the editor keeps the mode and text. Choose `OpenHands default` and Save (both fields `null`) before later checks that use this profile.- **Set as active (`F13.set-active`).** Run `control-openhands browser click 'testid=agent-profile-row >> has-text=QA_oh >> testid=agent-profile-menu-trigger'`, `control-openhands browser click 'testid=agent-profile-set-active'` and `control-openhands browser wait-text 'Switched to profile "QA_oh"' --timeout 10000`. After `browser reload`, `browser count 'testid=agent-profile-row >> has-text=QA_oh >> testid=agent-profile-active-badge'` and `browser count 'testid=agent-profile-active-badge'` are both `1`. Reopen the QA_oh menu: `browser enabled 'testid=agent-profile-set-active'` is `false` (screenshot `--feature F13.set-active --name menu`); `control-openhands browser press Escape` closes it (`browser count 'testid=agent-profile-actions-menu'` is `0`).-- **Pill wins over a pinned LLM (`F13.llm-pill-precedence`).** With `QA_oh` active (pinned to `deepseek-pro`) and the Home pill on `deepseek-flash` (`browser text 'testid=chat-input-llm-profile'` on `/`), run `control-openhands conversation start --prompt "Reply with only: qa-profile-ok" --wait --timeout 240`. The output's `model` is `deepseek/deepseek-flash`, and `control-openhands api GET /api/conversations/<id>` (id from that output) has `"launched_agent_profile": null` and `tool_concurrency_limit` `1`: the profile was not applied.-- **Active profile drives the launch (`F13.active-drives-conversation`).** Run `control-openhands browser goto /settings/agents` (the previous `conversation start` left the browser on the conversation), edit `QA_oh`, run `control-openhands browser choose 'testid=agent-profile-llm-selector' 'deepseek-flash (deepseek/deepseek-flash)'` (the toggle-button opener failed 5 of 5 times here: see Gotchas), check `browser value 'testid=agent-profile-llm-selector'`, Save (`Profile "QA_oh" updated`). Run the same `conversation start` again; `api GET /api/conversations/<id>` now has `launched_agent_profile.agent_profile_id` equal to QA_oh's id from `api GET /api/agent-profiles` and `tool_concurrency_limit` `2`; `control-openhands conversation events <id> --kinds MessageEvent` contains `qa-profile-ok`.-- **MCP scope (`F13.mcp-scope`).** With `qa_mcp_a` and `qa_mcp_b` arranged, `browser goto /settings/agents`, Edit `QA_oh`: `browser value 'testid=agent-settings-mcp-mode'` is `All servers` and the eval below (with `+(i.disabled?'(disabled)':'')` appended) shows both `=true(disabled)`. Pick `Choose servers` from `testid=agent-settings-mcp-mode`; `control-openhands browser eval "[...document.querySelectorAll('[data-testid=agent-settings-mcp-list] input')].map(i=>i.dataset.testid+'='+i.checked)"` shows both `=true`. Run `control-openhands browser click 'testid=agent-settings-mcp-list >> text=qa_mcp_b'` (now `false`), Save; `api GET /api/agent-profiles/QA_oh` has `mcp_server_refs` `["qa_mcp_a"]`. Dangling: arrange `control-openhands api DELETE /api/settings/mcp/qa_mcp_a --write`, `browser reload`, Edit `QA_oh`: `control-openhands browser text 'testid=agent-settings-mcp-dangling'` is `qa_mcp_a is no longer configured. Saving with it selected will stop this agent from starting — clear it or re-add the server.` (screenshot `--feature F13.mcp-scope --name dangling`). Leave without saving, `browser goto /`, `browser reload`, `control-openhands browser fill 'testid=chat-input' 'Reply with only: qa-dangling'`, `control-openhands browser click 'testid=submit-button' --observe 'role=status' --observe-ms 4000` and `control-openhands browser toasts --history` (QA_oh must still be active and pinned to the pill's LLM, as `F13.active-drives-conversation` leaves it; otherwise the launch skips the profile and nothing is refused): the launch is refused (one `POST /api/conversations` 422 in `browser errors --app-only`; earlier failures in the session are listed too): `browser toasts --history` shows `Creating conversation…`, then one toast carrying the server's `detail.message` (`MCP server(s) not configured: qa_mcp_a` on Agent Server 1.53.0), and `browser text 'testid=chat-input'` still reads `Reply with only: qa-dangling`. Clear the composer with `browser fill 'testid=chat-input' ''`, then clear the ref: `control-openhands browser goto /settings/agents`, Edit `QA_oh`, `browser click 'testid=agent-settings-mcp-list >> text=qa_mcp_a'` (the warning count drops to `0`), pick `All servers`, Save; `mcp_server_refs` is `null`.+- **Pill wins over a pinned LLM (`F13.llm-pill-precedence`).** With `QA_oh` active (pinned to `deepseek-pro`) and the Home pill on `deepseek-flash` (`browser text 'testid=chat-input-llm-profile'` on `/`), run `control-openhands conversation start --prompt "Reply with only: qa-profile-ok. Do not run tools." --wait --timeout 240`. The output's `model` is `deepseek/deepseek-flash`. On the current backend, `control-openhands api GET /api/conversations/<id>` (id from that output) has `launched_agent_profile.agent_profile_id` equal to QA_oh's id, `launched_agent_profile.llm_profile_ref` `deepseek-flash`, and `agent.tool_concurrency_limit` `2`: the per-launch LLM override keeps the named profile's other settings. These metadata checks can be made before a successful reply; a refused-loopback fixture proves the launch configuration only.+- **Active profile drives the launch (`F13.active-drives-conversation`).** Run `control-openhands browser goto /settings/agents` (the previous `conversation start` left the browser on the conversation), edit `QA_oh`, run `control-openhands browser choose 'testid=agent-profile-llm-selector' 'deepseek-flash (deepseek/deepseek-flash)'` (the toggle-button opener failed 5 of 5 times here: see Gotchas), check `browser value 'testid=agent-profile-llm-selector'`, Save (`Profile "QA_oh" updated`). Run the same `conversation start` again; `api GET /api/conversations/<id>` now has `launched_agent_profile.agent_profile_id` equal to QA_oh's id from `api GET /api/agent-profiles` and `agent.tool_concurrency_limit` `2`; `control-openhands conversation events <id> --kinds MessageEvent` contains `qa-profile-ok`.+- **MCP scope (`F13.mcp-scope`).** With `qa_mcp_a` and `qa_mcp_b` arranged, `browser goto /settings/agents`, Edit `QA_oh`: `browser value 'testid=agent-settings-mcp-mode'` is `All servers` and the eval below (with `+(i.disabled?'(disabled)':'')` appended) shows both `=true(disabled)`. Pick `Choose servers` from `testid=agent-settings-mcp-mode`; `control-openhands browser eval "[...document.querySelectorAll('[data-testid=agent-settings-mcp-list] input')].map(i=>i.dataset.testid+'='+i.checked)"` shows both `=true`. Run `control-openhands browser click 'testid=agent-settings-mcp-list >> text=qa_mcp_b'` (now `false`), Save; `api GET /api/agent-profiles/QA_oh` has `mcp_server_refs` `["qa_mcp_a"]`. Dangling: arrange `control-openhands api DELETE /api/settings/mcp/qa_mcp_a --write`, `browser reload`, Edit `QA_oh`: `control-openhands browser text 'testid=agent-settings-mcp-dangling'` is `qa_mcp_a is no longer configured. Saving with it selected will stop this agent from starting — clear it or re-add the server.` (screenshot `--feature F13.mcp-scope --name dangling`). Leave without saving, `browser goto /`, `browser reload`, `control-openhands browser fill 'testid=chat-input' 'Reply with only: qa-dangling'`, `control-openhands browser click 'testid=submit-button' --observe 'role=status' --observe-ms 4000` and `control-openhands browser toasts --history` (QA_oh must still be active; on older backends without per-launch LLM override it must also be pinned to the pill's LLM): the launch is refused (one `POST /api/conversations` 422 in `browser errors --app-only`; earlier failures in the session are listed too): `browser toasts --history` shows `Creating conversation…`, then one toast carrying the server's `detail.message` (`MCP server(s) not configured: qa_mcp_a` on Agent Server 1.53.0), and `browser text 'testid=chat-input'` still reads `Reply with only: qa-dangling`. Clear the composer with `browser fill 'testid=chat-input' ''`, then clear the ref: `control-openhands browser goto /settings/agents`, Edit `QA_oh`, `browser click 'testid=agent-settings-mcp-list >> text=qa_mcp_a'` (the warning count drops to `0`), pick `All servers`, Save; `mcp_server_refs` is `null`.- **Credential fields per preset (`F13.acp-preset-credentials`).** In the Add editor, `control-openhands browser choose 'testid=agent-type-selector' 'ACP (external subprocess)'`, then for each of `Codex`, `Gemini CLI` and `Custom` run `control-openhands browser choose 'testid=agent-preset-selector' '<preset>'` and `control-openhands browser testids --hidden --filter settings-acp`: Codex lists `settings-acp-secret-OPENAI_API_KEY` (input), `-CODEX_AUTH_JSON` (textarea), `-OPENAI_BASE_URL`; Gemini CLI lists `-GOOGLE_APPLICATION_CREDENTIALS_JSON` (textarea), `-GOOGLE_CLOUD_PROJECT`, `-GOOGLE_CLOUD_LOCATION`, `-GOOGLE_GENAI_USE_VERTEXAI`, `-GEMINI_API_KEY`, `-GEMINI_BASE_URL`; Custom lists nothing and `browser eval "document.body.innerText.includes('Credentials')"` is `false`. For Gemini, `control-openhands browser wait 'testid=settings-acp-auth-checking' --state detached --timeout 15000`, then `browser testids --hidden --filter settings-acp-auth` has count `0` (no login, no stored credential); `browser eval "document.querySelector('[data-testid=settings-acp-secret-GEMINI_API_KEY]').type"` is `password`, `GOOGLE_CLOUD_PROJECT` is `text`. Screenshot `--feature F13.acp-preset-credentials --name gemini` after `browser scroll 'testid=settings-acp-secret-GOOGLE_APPLICATION_CREDENTIALS_JSON'`. Cancel (Codex shows `settings-acp-auth-configured` instead while `OPENAI_API_KEY` from `F13.acp-credentials` is still saved).-- **Menu keyboard (`F13.menu-keyboard`).** Run `control-openhands browser goto /settings/agents`. It needs a non-active row: run `control-openhands browser click 'testid=add-agent-profile'`, `control-openhands browser fill 'testid=agent-profile-name-input' QA_kb`, `control-openhands browser click 'testid=save-agent-profile-btn'` and `control-openhands browser wait-text 'Profile "QA_kb" created' --timeout 10000`. Open its menu with `control-openhands browser click 'testid=agent-profile-row >> has-text=QA_kb >> testid=agent-profile-menu-trigger'` and right away run `control-openhands browser eval "document.activeElement?.dataset?.testid"`: `agent-profile-edit` (opening the menu moves focus to its first enabled item, as the LLM profile menu does in `F10.actions-menu`; fixed by #18112). Then `control-openhands browser press ArrowDown` followed by that eval four times: `agent-profile-set-active`, `agent-profile-delete`, `agent-profile-edit`, `agent-profile-set-active`; `browser press ArrowUp` returns to `agent-profile-edit`. `control-openhands browser press Escape` closes the menu (`control-openhands browser count 'testid=agent-profile-actions-menu'` is `0`) and returns focus to the row's own trigger (`control-openhands browser eval "(()=>{const a=document.activeElement;const r=a.closest('[data-testid=agent-profile-row]');return a.dataset.testid+'@'+(r?r.innerText.split('\\n')[0]:'-')})()"` is `agent-profile-menu-trigger@QA_kb`); reopen it and `control-openhands browser press Tab`: closed, and focus moves on past the trigger to the next row's trigger; keyboard open (`control-openhands browser focus` on the trigger, then `browser press Enter`) also lands on `agent-profile-edit`; reopen it and run `control-openhands browser mouse-click 900 600`: closed. Active row: open its menu with `control-openhands browser click '[data-testid=agent-profile-row]:has([data-testid=agent-profile-active-badge]) >> testid=agent-profile-menu-trigger'` (`control-openhands browser snapshot 'testid=agent-profile-actions-menu'` shows `menuitem "Set as active" [disabled]`; focus is already on `agent-profile-edit`), then `control-openhands browser press ArrowDown` and the eval, then `control-openhands browser press ArrowUp` and the eval: `agent-profile-delete` (the disabled **Set as active** is skipped), then `agent-profile-edit` (`control-openhands browser screenshot --feature F13.menu-keyboard --name active-row-menu`). At `browser viewport phone` the menu opens inside the viewport with the same focus order. With the menu still open, run `control-openhands browser focus 'testid=agent-profile-edit'` and `control-openhands browser press Enter`: `control-openhands browser text 'testid=agent-profile-editor-title'` is `Edit agent profile`. Leave the editor with `control-openhands browser click 'testid=cancel-agent-profile-btn'` (`browser count 'testid=add-agent-profile'` is `1`). Delete `QA_kb`: reopen its menu, `control-openhands browser click 'testid=agent-profile-delete'`, `control-openhands browser click 'testid=delete-agent-profile-confirm'` and `control-openhands browser wait-text 'Profile "QA_kb" deleted' --timeout 10000`.-- **Delete (`F13.delete`).** Run `control-openhands browser click 'testid=agent-profile-row >> has-text=QA_acp_renamed >> testid=agent-profile-menu-trigger'`, `control-openhands browser click 'testid=agent-profile-delete'`; `control-openhands browser text 'role=dialog'` reads `Delete Profile` and `Are you sure you want to delete the profile "QA_acp_renamed"? This action cannot be undone.` `control-openhands browser click 'role=dialog >> role=button[name="Cancel"]'` keeps the row (count `1`). Repeat and `control-openhands browser click 'testid=delete-agent-profile-confirm'`, `browser wait-text 'Profile "QA_acp_renamed" deleted' --timeout 10000`, `browser reload`: count `0`. Delete the active `QA_oh` the same way: after `browser reload`, `browser count 'testid=agent-profile-active-badge'` is `0` and `api GET /api/agent-profiles` has `active_agent_profile_id` `null`. Delete `QA_codex`, `QA_opencode`, `QA_pi` and `default`: the list again shows one row, `default` / `deepseek-flash` / `Default`, with a new id (the server re-seeds it).+- **Menu keyboard (`F13.menu-keyboard`).** Run `control-openhands browser goto /settings/agents`. It needs a non-active row: run `control-openhands browser click 'testid=add-agent-profile'`, `control-openhands browser fill 'testid=agent-profile-name-input' QA_kb`, `control-openhands browser click 'testid=save-agent-profile-btn'` and `control-openhands browser wait-text 'Profile "QA_kb" created' --timeout 10000`. Open its menu with `control-openhands browser click 'testid=agent-profile-row >> has-text=QA_kb >> testid=agent-profile-menu-trigger'` and right away run `control-openhands browser eval "document.activeElement?.dataset?.testid"`: `agent-profile-edit` (opening the menu moves focus to its first enabled item, as the LLM profile menu does in `F10.actions-menu`). Then `control-openhands browser press ArrowDown` followed by that eval four times: `agent-profile-set-active`, `agent-profile-delete`, `agent-profile-edit`, `agent-profile-set-active`; `browser press ArrowUp` returns to `agent-profile-edit`. `control-openhands browser press Escape` closes the menu (`control-openhands browser count 'testid=agent-profile-actions-menu'` is `0`) and returns focus to the row's own trigger (`control-openhands browser eval "(()=>{const a=document.activeElement;const r=a.closest('[data-testid=agent-profile-row]');return a.dataset.testid+'@'+(r?r.innerText.split('\\n')[0]:'-')})()"` is `agent-profile-menu-trigger@QA_kb`); reopen it and `control-openhands browser press Tab`: closed, and focus moves on past the trigger to the next row's trigger; keyboard open (`control-openhands browser focus` on the trigger, then `browser press Enter`) also lands on `agent-profile-edit`; reopen it and run `control-openhands browser mouse-click 900 600`: closed. Active row: open its menu with `control-openhands browser click '[data-testid=agent-profile-row]:has([data-testid=agent-profile-active-badge]) >> testid=agent-profile-menu-trigger'` (`control-openhands browser snapshot 'testid=agent-profile-actions-menu'` shows `menuitem "Set as active" [disabled]`; focus is already on `agent-profile-edit`), then `control-openhands browser press ArrowDown` and the eval, then `control-openhands browser press ArrowUp` and the eval: `agent-profile-delete` (the disabled **Set as active** is skipped), then `agent-profile-edit` (`control-openhands browser screenshot --feature F13.menu-keyboard --name active-row-menu`). At `browser viewport phone` the menu opens inside the viewport with the same focus order. With the menu still open, run `control-openhands browser focus 'testid=agent-profile-edit'` and `control-openhands browser press Enter`: `control-openhands browser text 'testid=agent-profile-editor-title'` is `Edit agent profile`. Leave the editor with `control-openhands browser click 'testid=cancel-agent-profile-btn'` (`browser count 'testid=add-agent-profile'` is `1`). Delete `QA_kb`: reopen its menu, `control-openhands browser click 'testid=agent-profile-delete'`, `control-openhands browser click 'testid=delete-agent-profile-confirm'` and `control-openhands browser wait-text 'Profile "QA_kb" deleted' --timeout 10000`.+- **Delete (`F13.delete`).** Run `control-openhands browser click 'testid=agent-profile-row >> has-text=QA_acp_renamed >> testid=agent-profile-menu-trigger'`, `control-openhands browser click 'testid=agent-profile-delete'`; `control-openhands browser text 'role=dialog'` reads `Delete Profile` and `Are you sure you want to delete the profile "QA_acp_renamed"? This action cannot be undone.` `control-openhands browser click 'role=dialog >> role=button[name="Cancel"]'` keeps the row (count `1`). Repeat and `control-openhands browser click 'testid=delete-agent-profile-confirm'`, `browser wait-text 'Profile "QA_acp_renamed" deleted' --timeout 10000`, `browser reload`: count `0`. Delete the active `QA_oh` the same way: after `browser reload`, `browser count 'testid=agent-profile-active-badge'` is `1`, the `default` row has that badge, and `api GET /api/agent-profiles` has `active_agent_profile_id` equal to default's id. Delete `QA_codex`, `QA_opencode`, `QA_pi` and `default`: the list again shows one row, `default` / `deepseek-flash` / `Default`, with a new id (the server re-seeds it).- **Phone (`F13.phone`).** Run `control-openhands browser viewport phone`, `control-openhands browser goto /settings/agents`, `control-openhands browser bbox 'testid=agent-profile-row'` (`insideViewport` `true`, `pageHorizontalOverflow` `false`) and `browser screenshot --feature F13.phone --name list`. Open the row menu: `browser bbox 'testid=agent-profile-actions-menu'` is inside the viewport. `browser press Escape`, click `testid=add-agent-profile`: `browser bbox 'testid=agent-settings-screen'` is 348 px wide at x 16 with no page overflow (`insideViewport` is `false` only because the form is taller than the screen). Switch to ACP, `browser scroll 'testid=settings-acp-secret-ANTHROPIC_API_KEY'` and screenshot `--name acp-credentials`. Cancel and `control-openhands browser viewport desktop`.- **Errors.** After the family, `control-openhands browser errors --app-only` shows no page errors and no `[i18n] Missing translation` warnings (the Parallel tool calls label and description are translated since #18035), only the deliberate 422s from the dangling-MCP launch.- **Blocked (`F13.acp-conversation`, `F13.cloud-read-only`).** Record `control-openhands evidence add --feature F13.acp-conversation --result blocked ...` naming the missing provider CLI/account, and the same for `F13.cloud-read-only` naming the Cloud member login.- **Cleanup.** Delete the `OPENAI_API_KEY` credential through Secrets: `control-openhands browser goto /settings/secrets`, `control-openhands browser click 'testid=secret-item >> has-text=OPENAI_API_KEY >> testid=delete-secret-button'`, `control-openhands browser click 'testid=confirmation-modal >> testid=confirm-button'`. Remove `qa_mcp_b` with `control-openhands api DELETE /api/settings/mcp/qa_mcp_b --write`. Leave `default` active.+- **Default wording (`F13.list`, `F13.set-active`).** Run `control-openhands browser goto /settings/agents`, `control-openhands browser text 'testid=agent-profile-active-badge'`, open the active row's `testid=agent-profile-menu-trigger`, and `control-openhands browser snapshot 'testid=agent-profile-actions-menu'`. Expected: the menu uses the same Default wording as the badge. Known failure, reproduced 2026-10-12 ([#17426](https://github.com/OpenHands/OpenHands/issues/17426)): the badge reads `Default` and the menu reads `Set as active`. Close with Escape, reopen once to repeat, then Escape; no activation or model call is required.++- **Seeded default key propagation (`F13.active-drives-conversation`).** With the seeded `default` agent profile active and the disposable `QA_LOOPBACK` LLM profile active, `control-openhands api GET /api/settings --pick llm_api_key_is_set` is `true`. Run `control-openhands conversation start --prompt 'Reply with exactly pong. Do not run tools.' --timeout 60` (the dummy endpoint refuses connections). Read only boolean presence of `agent.llm.api_key` in the run-owned persisted `base_state.json`, keeping the value private, and inspect `conversation status <id>`. Expected: the key is populated and initialization reaches the refused loopback endpoint rather than an absent-key error. This checks propagation before provider authentication or a successful reply.++- **Gemini auth file is not sufficient (`F13.acp-credentials`).** In the isolated run HOME only, create an empty `{}` file at `private/home/.gemini/oauth_creds.json` (never replace a real host login). Open Add agent profile → ACP → Gemini CLI and read `testid=settings-acp-auth-detected`; Cancel and repeat a fresh Add form. Expected: no authenticated banner for an empty file. Known failure, reproduced twice 2026-10-12 ([OpenHands/software-agent-sdk#4629](https://github.com/OpenHands/software-agent-sdk/issues/4629)): `You're already signed in to Gemini CLI — you can leave the fields below blank.` No provider or model call is needed. Remove the own dummy auth file; SDK auth-method priority and upstream CLI-version branches remain outside this check.++- **Published stale ACP binary on PATH (`F13.acp-conversation`).** Install the real published `@zed-industries/[email protected]` under the owned run's `workspace/qa-stale-adapter` (no global install). Create/activate a real Codex ACP profile through Settings > Agents with only a dummy API-key presence marker and loopback Base URL. Prepend that package's `node_modules/.bin` only during the own backend `restart`; keep its private HOME, set default OpenAI routes plus HTTP(S)/ALL proxy to refused loopback and `NO_PROXY=127.0.0.1,localhost`, with cached dependencies offline. Doctor, then start two actual Home chats and inspect the sanitized runtime command and initialization version. Expected: the SDK uses the pinned `1.10.0` adapter. Known failure, reproduced twice 2026-10-12 ([OpenHands/software-agent-sdk#5466](https://github.com/OpenHands/software-agent-sdk/issues/5466)): `acp_command` became bare `codex-acp`, and actual ACP initialization reported `agent_version=0.16.0`; the version warning still wrongly blamed npx fallback. Both then stopped at `Authentication required`. This proves selection/warning before authentication, not a usable session, the historical model-cache error or the Claude counterpart. Restore the launcher environment/PATH and default agent, delete the own profile/dummy secrets/package fixture, then doctor and stop.++- **Suffixed profile keeps its identity (`F13.edit`).** A normal edit preserves the profile UUID and increments its revision. Known failure (reproduced 2026-10-16, [OpenHands/software-agent-sdk#5873](https://github.com/OpenHands/software-agent-sdk/issues/5873)): two real concurrency-only saves of a `.json` name each minted a new UUID at revision0. SDK1.55 repaired the active pointer to default; do not claim it remained dangling. The credential-free block checks the revision symptom and cleans its own fixtures.++ ```sh known-bug F13.edit json-suffix+ trap 'control-openhands api DELETE /api/agent-profiles/QA_KNOWN_SUFFIX --write >/dev/null' EXIT+ control-openhands browser viewport desktop+ control-openhands browser goto /settings/agents+ control-openhands browser click 'testid=add-agent-profile'+ control-openhands browser fill 'testid=agent-profile-name-input' QA_KNOWN_SUFFIX.json+ control-openhands browser click 'testid=save-agent-profile-btn'+ control-openhands browser wait-text 'Profile "QA_KNOWN_SUFFIX.json" created' --timeout 10000+ control-openhands assert --field value --equals 0 -- api GET /api/agent-profiles/QA_KNOWN_SUFFIX --pick profile.revision+ control-openhands browser click 'testid=agent-profile-row >> has-text=QA_KNOWN_SUFFIX >> testid=agent-profile-menu-trigger'+ control-openhands browser click 'testid=agent-profile-edit'+ control-openhands browser fill 'testid=sdk-settings-tool_concurrency_limit' 2+ control-openhands browser click 'testid=save-agent-profile-btn'+ control-openhands browser wait-text 'Profile "QA_KNOWN_SUFFIX.json" updated' --timeout 10000+ control-openhands assert --field value --equals 1 --bug-shows 0 -- api GET /api/agent-profiles/QA_KNOWN_SUFFIX --pick profile.revision # bug+ ```+## Gotchas- The editor is not a route: `browser url` stays `/settings/agents` and `browser reload` drops unsaved edits back to the list. The settings page header (`settings-page-subtitle`) is hidden while it is open.-- Autocomplete dropdowns: clicking an input that already shows a value can open an empty listbox (the text filters it), and an option click then times out. The toggle-button opener does not always open the list either: on the Edit editor's `agent-profile-llm-selector` (bottom of the form) the option click timed out 5 of 5 times right after opening the editor, and the model list's `Custom` option once detached mid-click after a preset change. `browser choose` was reliable on every dropdown. A listbox also closes if you spend a command (snapshot, eval) in between, so click the option right after opening.+- Autocomplete dropdowns: clicking an input that already shows a value can open an empty listbox (the text filters it), and an option click then times out. The toggle-button opener does not always open the list either: on the Edit editor's `agent-profile-llm-selector` (bottom of the form) the option click timed out 5 of 5 times right after opening the editor, and the model list's `Custom` option once detached mid-click after a preset change. `browser choose` avoids the input-filtering problem, but an option may still remain unstable immediately after opening an editor. Reload and reopen before retrying; the suffix-identity check can use the already-selected active LLM without changing that dropdown. A listbox also closes if you spend a command (snapshot, eval) in between, so click the option right after opening.- A taken name marks the field invalid with an alert reading `An agent profile with this name already exists. …`.- `has-text=default` also matches the active row's `Default` badge (Playwright's text match ignores case), so with several profiles `'testid=agent-profile-row >> has-text=default'` hits two rows; select the seeded profile with `'testid=agent-profile-row >> nth=-1'` (it sorts last) or by its badge.- Model lookups (`F13.acp-model-discovery`) spawn the provider CLI on the agent-server machine through npx with the profile's allowed secrets; they send no prompt. A failed lookup is retried each time the preset is picked again; a successful one is cached for about five minutes per preset and secret set.- Switch rows are `sr-only` checkboxes inside labels (keyboard reachable): click the label text (`'testid=agent-settings-tool-list >> text=<tool>'`, `'testid=agent-settings-mcp-list >> text=<name>'`) and read `.checked` with `browser eval`. A bare `label` selector segment is parsed as `label=`, not CSS.- The Default badge, the **Set as active** item and the `Switched to profile` toast all refer to the same active pointer; the wording differs from the LLM page (**Set as default**).- The OpenHands row's grey text is the LLM profile name, not the kind. A fresh run's `default` shows `default` there although no LLM profile of that name exists until it is edited (`F13.stale-llm-ref`).-- Home launches follow the LLM pill: a named profile pinned to another LLM profile launches through the legacy settings path, so its tool, concurrency and MCP settings do not apply (`F13.llm-pill-precedence`). To prove a profile's settings, pin it to the active LLM profile, and read `launched_agent_profile` on the conversation.-- The ACP auth probe runs the provider CLI on the agent-server machine. On this sandbox the host's Claude CLI is logged in, so Claude Code shows "already signed in". Never launch an ACP conversation on that login.+- Local seeded OpenHands `default` launches through global agent settings so the Skills selection applies; its per-profile fields do not apply on that path. Named profiles use their saved settings. Home launches follow the LLM pill. On a backend advertising per-launch LLM override, a named profile pinned to another LLM keeps its tool, concurrency and MCP settings; read `launched_agent_profile` and `agent` on the conversation (`F13.llm-pill-precedence`). Only older backends without that capability use the legacy settings fallback when the profile and pill differ.+- The ACP auth probe runs the provider CLI on the agent-server machine. If that machine has a Claude CLI login, Claude Code can show "already signed in". Use only an explicitly approved account for authenticated ACP checks.- ACP credentials are global secrets, not profile fields: cancelling the editor saves nothing, but saving any ACP profile with a typed credential writes it to Settings → Secrets (F14) for every conversation.- **Choose secrets** on an ACP profile also selects provider secret names that are not saved (`CODEX_AUTH_JSON`, `OPENAI_BASE_URL`); they show as switches without descriptions.- The empty state (`agent-profiles-empty`) and the load-error state (`agent-profiles-load-error`) are not reachable on a local backend: deleting the last profile re-seeds `default`, and with the agent-server stopped the app swaps to **Manage backends** after a reload, or keeps the cached list without one.- The refused-launch toast text comes from the Agent Server and changes between versions (`MCP server(s) not configured: <name>` on 1.53.0): assert that one toast names the server and the prompt stays, not the exact wording.-- Row-menu keyboard: #18112 (closed #18060) ported #18019's LLM-menu fix: opening focuses the first enabled item, arrows skip the disabled **Set as active**, and Escape/Tab return focus to the trigger first (verified 2026-10-13 on desktop and phone). A checkout without it leaves focus on the trigger and ArrowDown stuck on the active row's Edit.-- #18035 (closed #17939) gave the Add editor its own description and translated the Parallel tool calls field. A checkout without it shows the LLM page's `Configure your LLM settings below, then click Save to create this profile.` and logs `[i18n] Missing translation for key "SCHEMA$TOOL_CONCURRENCY_LIMIT$LABEL"` / `$DESCRIPTION` whenever the editor renders.+- Opening a row menu focuses its first enabled item. Arrows skip disabled **Set as active**; Escape closes the menu and returns focus to its trigger, while Tab closes it and continues normal tab order.+- The Add agent editor has its own agent description. Parallel tool calls label and description are translated.+- Known issue [#16905](https://github.com/OpenHands/OpenHands/issues/16905): An enabled built-in OpenHands skill can be absent from a new Claude Code ACP session. The enabled/disabled instruction-delivery check needs an approved authenticated Claude Code account, its ACP subprocess and the configured Code Review automation; no provider session is authorized here. (`F13.acp-conversation`).+- Known issue [#16296](https://github.com/OpenHands/OpenHands/issues/16296): Claude Code credential fields remain visible below the already-connected banner, leading users to think keys are required. The connected-state check needs an approved existing Claude CLI login on the agent-server host; a dummy credential does not prove that signed-in state. (`F13.acp-credentials`).+- Known issue [#17051](https://github.com/OpenHands/OpenHands/issues/17051): Codex ACP can work in New Chat while Issue-to-PR automation fails with missing OpenAI API credentials. Reproduction needs an approved Codex subscription login plus a configured Issue-to-PR automation and its repository credentials; the report does not identify the saved automation/profile settings. (`F13.acp-conversation`).+- Known issue [OpenHands/software-agent-sdk#5167](https://github.com/OpenHands/software-agent-sdk/issues/5167): On a headless Docker host an already usable Codex ChatGPT login can still time out during ACP authenticate. Reproduction needs an approved ChatGPT login in the reported headless/read-only-auth-file deployment and its adapter; a desktop or bogus login cannot establish this conditional trigger. (`F13.acp-conversation`).+- Known issue [OpenHands/software-agent-sdk#5995](https://github.com/OpenHands/software-agent-sdk/issues/5995): Claude Code ACP first turns with `claude-opus-5-5` can fail because the pinned adapter bundles an older Claude Code. Blocked: needs an approved Claude login/API key entitled to Opus5.5 and a real first turn; current SDK1.55 still pins the reported adapter0.63.0. No authenticated turn was attempted. (`F13.acp-conversation`, `F13.acp-model-discovery`).+- Known issue [OpenHands/software-agent-sdk#5899](https://github.com/OpenHands/software-agent-sdk/issues/5899): The browser tool can be reported unusable when genuine Chromium exists only under `PLAYWRIGHT_BROWSERS_PATH`. Blocked on this host: it has standard user Chrome. The distinguishing fixture needs a host/container with no standard/default-cache/PATH browser, plus a genuine custom-cache Chromium; never remove the user browser to create it. (`F13.openhands-options`; operational browser path F27).diff --git a/.agents/skills/verify-openhands/references/feature-map/F14-secrets.md b/.agents/skills/verify-openhands/references/feature-map/F14-secrets.mdindex dcd2126da0..b63e379aca 100644--- a/.agents/skills/verify-openhands/references/feature-map/F14-secrets.md+++ b/.agents/skills/verify-openhands/references/feature-map/F14-secrets.md@@ -66,6 +66,8 @@ Preconditions:```- **Escape cancels delete (`F14.delete-escape`).** Run `control-openhands browser goto /settings/secrets`, `control-openhands browser click 'testid=secret-item >> has-text=QA_RENAMED_SECRET >> testid=delete-secret-button'`, `control-openhands browser press Escape`, then `control-openhands browser count 'testid=confirmation-modal'` (`0`) and `browser count 'testid=secret-item >> has-text=QA_RENAMED_SECRET'` (`1`). Clean up: delete `QA_RENAMED_SECRET` and then `QA_REQ_SECRET` (from `F14.agent-request`), each with `delete-secret-button` → `testid=confirmation-modal >> testid=confirm-button`, `browser reload`; `api GET /api/settings/secrets` lists only `OPENHANDS_AUTOMATION_API_KEY`.+- **Cipher mismatch refuses mutation (`F14.create`, `F14.edit`, `F14.delete`).** This is an external-state fixture, not a credential test. Arrange two own dummy secrets and a dummy loopback LLM key; privately back up the run-owned `encryption-key`, `settings.json` and `secrets.json`. Replace only the own cipher file with a newly generated value (never display it), restart and doctor. After bootstrap, restore the old encrypted store bytes under the replacement cipher to isolate the UI write. Add a new dummy secret through the real Secrets form, and inspect only boolean preservation metadata / availability of the old dummy value privately. Independently restore the encrypted settings baseline and change Sound Notifications through Application → Save. Expected: reject writes when existing encrypted state cannot decrypt, preserving it. Known failure, reproduced 2026-10-12 ([OpenHands/software-agent-sdk#5622](https://github.com/OpenHands/software-agent-sdk/issues/5622)): the secret create succeeds and rewrites an old entry while its dummy value is unavailable (getter 404, name still listed); two unrelated settings saves persist the old LLM key as null. Repeat with a fresh replacement cipher. Immediately restore the original cipher and both stores in `finally`, restart and doctor; original state must be readable again. This drives secret creation and settings Save, not every edit/delete permutation. Keep all cipher bytes, values and raw store snapshots private.+## Gotchas- Settings nav test ids embed the route, slashes included: `sidebar-settings-/settings/secrets`. Quote the selector as written.@@ -80,3 +82,4 @@ Preconditions:- `request_secret` is registered on every new OpenHands-agent conversation this frontend launches (not on ACP agents or older conversations) and appears in Show Agent Tools & Metadata (see `F07.agent-tools-modal`). The agent learns the outcome from a hidden `[secret-request] ...` user message; read it with `conversation events <id> --kinds MessageEvent`, not from the chat.- At 390 px the name column truncates long names (`OPEN…`); use `browser text` on the row rather than the screenshot to read a name.- `OPENHANDS_AUTOMATION_API_KEY` is seeded by the launcher for automations; deleting it breaks F18/F19 runs in the same stack.+- Known issue [#17710](https://github.com/OpenHands/OpenHands/issues/17710): On the Cloud backend, deleting a secret reportedly adds a suffixed duplicate and leaves the original. Reproduction needs a disposable writable OpenHands Cloud account at the affected service; Local secret CRUD exercises a different transport. (`F14.delete`, `F14.edit`).diff --git a/.agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md b/.agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.mdindex 234a3ec136..86b6ee1daf 100644--- a/.agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md+++ b/.agents/skills/verify-openhands/references/feature-map/F15-agent-behavior-settings.md@@ -78,6 +78,14 @@ Preconditions:- **Schema unavailable (`F15.schema-unavailable`, blocked).** `control-openhands service stop agent-server` does not reach this state: after `browser reload` the whole app is replaced by the **Manage backends** dialog (`Disconnected`, 502). Bring the stack back with `control-openhands restart`. The state needs a backend whose schema endpoints 404 (see Preconditions); there `browser text 'testid=sdk-schema-unavailable'` should start with `SDK settings schema unavailable.`- **Errors sweep.** `control-openhands browser errors --clear`, open the three pages and their All tabs, then `control-openhands browser errors --app-only`: no page errors and no warnings (the condenser fields are translated since #18035).+- **Background refetch keeps the draft (`F15.save-dirty-state`).** With Max Size stored as `240`, run `control-openhands browser goto /settings/condenser`, `control-openhands browser click 'testid=sdk-section-all-toggle'` and `control-openhands browser fill 'testid=sdk-settings-condenser.max_size' 111`. Arrange a changed server baseline with `control-openhands api PATCH /api/settings --write --data '{"agent_settings_diff":{"condenser":{"max_size":300}}}'`; this API write is a precondition, not the UI proof. Run `control-openhands browser clock --offset-ms 360000` (makes the mounted query stale), `control-openhands browser network-state offline`, `control-openhands browser network-state online` and `control-openhands browser network --filter '/api/settings'`. After the real GET refetch, the unsaved `111` should remain and Save should stay enabled. Known failure (reproduced 2026-10-12, [#16948](https://github.com/OpenHands/OpenHands/issues/16948)): `control-openhands browser value 'testid=sdk-settings-condenser.max_size'` is `300` and `control-openhands browser enabled 'testid=save-button'` is `false`; the draft vanished without Save, reset or a scope change. Capture `control-openhands browser screenshot --feature F15.save-dirty-state --name refetch-discards-draft`. Restore with `control-openhands browser clock --offset-ms 0`, fill Max Size `240`, click Save and confirm `control-openhands api GET /api/settings --pick agent_settings.condenser.max_size` is `240`; networking stays online.++- **Negative condenser input floor (`F15.save-errors`).** On `/settings/condenser` select All, fill `testid=sdk-settings-condenser.max_size` with `-5` and read `browser enabled 'testid=save-button'`. Expected by [#15813](https://github.com/OpenHands/OpenHands/issues/15813): the input prevents values below zero. Known failure, reproduced 2026-10-12: `-5` remains and Save is enabled. Save shows `Settings validation failed`, while `api GET /api/settings --pick agent_settings.condenser.max_size` remains `240`; repeat filling `-5` and read Save enabled, then fill `240` to discard the invalid draft. The server guard works; the requested client input floor does not.++- **Default Verification tabs (`F15.verification-page`, `F15.verification-critic-fields`).** [#15811](https://github.com/OpenHands/OpenHands/issues/15811) asks whether Basic and Advanced should differ. On defaults, snapshot `testid=verification-settings-screen` in each tab: both show only Confirmation Mode and Enable Critic. Enable Confirmation Mode without saving; `browser count 'testid=sdk-settings-security_analyzer'` is `1` on Advanced, `0` on Basic and `1` again on Advanced. Reload to discard, then repeat the default/conditional check. Verified 2026-10-12: the default sameness is the current conditional-field behavior, not evidence that the two tiers always render the same fields.++- **Datetime refreshed on Canvas launch (`F15.agent-context-page`).** Before a new chat, read `api GET /api/settings --pick agent_settings.agent_context.current_datetime`; record this public timestamp so it can be restored. Arrange an old persisted-state timestamp through the real API (`api PATCH /api/settings --write --data '{"agent_settings_diff":{"agent_context":{"current_datetime":"2001-01-01T00:00:00+00:00"}}}'`). Start a chat through the Home composer with the dummy loopback profile, then inspect the run-owned runtime `agent.agent_context.current_datetime` and `conversation events <id> --kinds SystemPromptEvent --grep 2001-01-01 --from-start --last 1`. Expected: runtime datetime matches the new launch and the old-date event search returns zero. Restore the original settings timestamp through PATCH; no successful model reply is required to inspect initialization.+## Gotchas- Switches are `sr-only` checkboxes (keyboard reachable): toggle by clicking the label text scoped to the page, read `.checked` with `browser eval`.@@ -87,7 +95,7 @@ Preconditions:- The page opens on the least detailed tier that shows every stored override (any minor override → All). Do not assume Basic after a reload.- Condenser Kind **None** (`no_op`) still shows the summarizer fields with defaults, although the stored condenser no longer has them, and edits to them are accepted (`Settings saved`) but dropped while **None** is stored. #18049 covers hiding fields that do not apply to the selected kind.- A server-side validation error is retried three times (about 1.5 s of `Saving...`) before the toast, and the toast only says `Settings validation failed` without naming the field (#17932).-- Condenser Kind is served by the Agent Server's settings schema as a user field. Since #18035 (closed #17939) Canvas shows its own choice labels and help text instead of the raw `llm_summarizing`/`no_op` values and the schema's reStructuredText description. It stays a user field on purpose: software-agent-sdk#5531, which dropped it from the schema, was closed because more condenser kinds may come, so the dropdown (and `F15.condenser-kind`) is expected behavior and its presentation is Canvas's job. OpenHands/software-agent-sdk#5495 (the schema's developer-facing description) is still open.+- Condenser Kind is served by the Agent Server's settings schema as a user field. Canvas shows its own choice labels and help text rather than raw `llm_summarizing`/`no_op` values or reStructuredText. The dropdown (`F15.condenser-kind`) remains a user setting so additional kinds can be selected. [OpenHands/software-agent-sdk#5495](https://github.com/OpenHands/software-agent-sdk/issues/5495), the schema's developer-facing description, remains an upstream issue.- Toasts close after a few seconds; read them right after the click, or assert the stored value with `api GET /api/settings`.- Under confirmation mode the first confirmed command sometimes returned `The terminal session was reset because the underlying tmux server/session disappeared...` while several runs shared one tmux server (OpenHands/OpenHands#17946); `launch` now isolates them, so treat a reset as a finding. The agent then asks to run it again; a confirmed command took up to two minutes to produce its observation. Confirm again and wait with `conversation wait`, not a fixed sleep.- The confirmation buttons have test ids (`action-confirm-button`, `action-reject-button`) but their accessible names are `Confirm action` / `Reject action`, not the visible `Continue` / `Cancel`: `role=button[name*="Continue"]` finds nothing.diff --git a/.agents/skills/verify-openhands/references/feature-map/F16-application-settings.md b/.agents/skills/verify-openhands/references/feature-map/F16-application-settings.mdindex ce13d1d90f..d43dd29835 100644--- a/.agents/skills/verify-openhands/references/feature-map/F16-application-settings.md+++ b/.agents/skills/verify-openhands/references/feature-map/F16-application-settings.md@@ -102,6 +102,8 @@ Preconditions:- **Backend switch in place (`F16.backend-switch`).** Needs a second backend with a different Git username: the second stack from F25's Preconditions added as `QA_Second` (F25 Add bullet), arranged with `control-openhands api PATCH /api/settings --run "$OH_VERIFY_RUN_2" --write --data '{"misc_settings_diff":{"app_preferences":{"git_user_name":"QA Second Git"}}}'`. With Local active, run `control-openhands browser goto /settings/app` and `control-openhands browser reload`; `browser value 'testid=git-user-name-input'` is `openhands`. Switch without leaving the page: `control-openhands browser click 'testid=backend-selector'`, `control-openhands browser click 'testid=backend-selector >> role=option[name="Connected QA_Second"]'`, `control-openhands browser wait 'testid=environment-switch-overlay' --state detached --timeout 10000`, wait a few seconds: the value is `QA Second Git` and `browser text 'testid=backend-synced-settings-badge'` names QA_Second (a first visit to a backend loads fresh). Switch back the same way to `Connected Local`: the value is `openhands` beside `synced from Local backend`, and it keeps following the active backend on further switches. The form remounts per backend, so an unsaved edit is dropped: on Local run `control-openhands browser fill 'testid=git-user-name-input' 'QA Local Edit'` (`browser enabled 'testid=submit-button'` is `true`), switch to `QA_Second` (`QA Second Git`, Save `false`) and back (`openhands`, Save `false`); `control-openhands api GET /api/settings --pick misc_settings.app_preferences.git_user_name` is unchanged. At `browser viewport phone` the selector is in the drawer (`testid=sidebar-mobile-menu-toggle`, then `'testid=sidebar-mobile-drawer >> testid=backend-selector'` and `role=option[name="Connected QA_Second"]`; close the drawer with `browser mouse-click 370 400`) and the field follows the same way. Switch back to Local and remove `QA_Second` afterwards (F25).- After the family, `control-openhands browser errors --app-only` reports `appErrors: 0`. Requests to `registry.npmjs.org` (update check) and Google Fonts fail through the sandbox proxy; they are not app errors.+- **Hard Do Not Track (`F16.analytics`).** With the harness build's `VITE_DO_NOT_TRACK=1`, open `/settings/app`, read `browser network --external --filter 'posthog|z.openhands.dev'` and `browser storage`. Grant analytics through the real switch and Save, reload, inspect those again, then revoke and Save to restore. Expected: zero PostHog requests and no PostHog storage key, including after opt-in. This checks the build flag; browser-only DNT and builds without it have separate paths.+## Gotchas- Two save models on one form: theme, checklist and voice input apply instantly and ignore Save; the rest is dropped silently if you navigate away before Save (filling `git-user-name-input`, then clicking `sidebar-settings-/settings/secrets` and back, shows the stored `openhands` again; `browser dialogs` is empty). Assert the instant ones without clicking Save, and never expect Save to enable for them.diff --git a/.agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md b/.agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.mdindex f560e8615e..0ecc62b391 100644--- a/.agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md+++ b/.agents/skills/verify-openhands/references/feature-map/F17-mcp-servers.md@@ -39,7 +39,7 @@ Source: `src/routes/extensions-hub.tsx`, `src/routes/mcp.tsx`, `src/components/f- `F17.custom-edit`: clicking a card opens **Edit MCP server** prefilled (secrets shown as `**********`); Save tests and persists the change.- `F17.stored-secret-reuse`: a saved server's env values and tokens come back from the settings API and in the editor as `**********`; the editor's Test connection swaps each placeholder for the stored encrypted value (never the placeholder text, never the plaintext) and Save leaves untouched secrets out of its sparse patch, so a Save from the edit form that keeps them changes only the edited fields.- `F17.single-request-mutations`: adding, editing or deleting one server sends exactly one `POST`, `PATCH` or `DELETE` to `/api/settings/mcp/<name>` and leaves sibling servers and their stored credentials untouched.-- `F17.credential-probe`: a catalog entry whose server lists its tools with any credentials (Slack) also gets a read-only credential probe during Install and Test connection: a failed probe reads `Credential check failed: <provider error>` and saves nothing (Known failure, reproduced 2026-10-08: any errored probe call, a network `fetch failed` included, is worded as a credential failure; #18161). A hosted server that rejects the token at connect (GitHub) never reaches a probe and reads `Connection failed: … 401 Unauthorized …`, also saving nothing (Known failure, reproduced 2026-10-08: the Agent Server's `POST /api/mcp/test` returns the 401 as `error_kind` `unknown`, so the modal cannot call it a credential failure; OpenHands/software-agent-sdk#5607); a server the catalog does not know gets no probe.+- `F17.credential-probe`: a catalog entry whose server lists its tools with any credentials (Slack) also gets a read-only credential probe during Install and Test connection: a failed probe reads `Credential check failed: <provider error>` and saves nothing (Known failure, reproduced 2026-10-08: any errored probe call, a network `fetch failed` included, is worded as a credential failure; #18161). A hosted server that rejects the token at connect (GitHub) never reaches a probe and reads `Connection failed: … 401 Unauthorized …`, also saving nothing (Known failure, reproduced 2026-10-08: the Agent Server's `POST /api/mcp/test` returns the 401 as `error_kind` `unknown`, so the modal cannot call it a credential failure; [OpenHands/software-agent-sdk#5607](https://github.com/OpenHands/software-agent-sdk/issues/5607)); a server the catalog does not know gets no probe.- `F17.delete-server`: **Delete** in the editor asks for confirmation; Cancel keeps the server, Confirm removes it with `MCP server removed.`- `F17.native-git-tabs`: on a Cloud backend, GitHub/GitLab/Bitbucket offer a recommended native-integration tab next to the MCP tab (blocked here).- `F17.agent-uses-server`: a conversation started after the save can call the server's tools.@@ -101,9 +101,17 @@ Preconditions:- **Cloud backend (`F17.cloud-nav`, `F17.native-git-tabs`).** Blocked without an OpenHands Cloud backend. On the local backend `control-openhands browser click 'testid=mcp-marketplace-card-github'` shows only the MCP form (`browser count 'testid=mcp-install-tab-native'` is `0`). With Cloud active, `browser text 'testid=extensions-navbar-desktop'` should list only MCP Servers and the cloud Skills link (`browser attr 'testid=sidebar-extensions-/skills' href` ends in `/settings/skills`), and the GitHub modal should show `mcp-install-tab-native` (Recommended) and `mcp-install-tab-mcp`.- **Cleanup.** On `/mcp` (not the conversation page `conversation start` left open), delete `qa_time`, `qa_shttp`, `qa_vault` and any `time`, `time_1`, `tavily`, `filesystem` cards through the editor's Delete: for each id, `browser focus '[data-server-id=<id>]'`, `browser press Enter`, `browser click 'testid=mcp-custom-editor-delete'`, `browser click 'testid=confirmation-modal >> testid=confirm-button'`, `browser wait '[data-server-id=<id>]' --state detached --timeout 10000`. After `browser reload`, `browser count 'testid=mcp-installed-empty'` is `1` and `api GET /api/settings --pick agent_settings.mcp_config` is `{}`. Delete the `TAVILY_API_KEY` secret on `/settings/secrets` (F14: `testid=secret-item >> has-text=TAVILY_API_KEY >> testid=delete-secret-button`, then `testid=confirmation-modal >> testid=confirm-button`), stop the Streamable HTTP fixture by its PID, and restore `deepseek-flash` as the active profile if you changed it.+- **Health belongs to a backend (`F17.server-health`).** Arrange a real `fixture mcp-server` echo entry. Register a second disposable backend identity against the own stack's direct Agent Server port (the first uses its ingress); this exercises separate registry identities with one physical stack. On Local open `/mcp`, probe the echo server and wait for `Reachable — credentials not verified`. Switch through `backend-selector` → `role=listbox >> text="QA_SCOPE"` and read the same card without `browser goto` or reload (both clear the global store). Expected: the unprobed identity is unchecked. Known failure, reproduced twice 2026-10-12 ([#16954](https://github.com/OpenHands/OpenHands/issues/16954)): the new identity shows the first identity's Reachable verdict. Return to Local, reload only before the next independent trial, probe and switch again. This proves the identity collision; different credentials, delayed probes and revision rotation on independent servers are additional branches. Restore Local and remove the alias afterwards.++- **Incomplete sibling must not hide working tools (`F17.custom-add`, `F17.agent-uses-server`).** Use `fixture mcp-server --name qa-settings-echo` and the real settings API to store its printed stdio command/args under `qa_settings_echo`. Probe it through the MCP card; a UI-started dummy-loopback conversation's `SystemPromptEvent` includes `qa_echo`. Arrange `api POST /api/settings/mcp/qa_empty --write --data '{}'`. Expected: 422 and no entry; valid tools remain available. Known failure, reproduced twice 2026-10-12 ([OpenHands/software-agent-sdk#5643](https://github.com/OpenHands/software-agent-sdk/issues/5643)): 201 stores `{"enabled":true}`, the entry has no card in the UI, and a real Home-composer start initializes without the valid echo tool. Delete `qa_empty` and recheck the good-only baseline. No successful model reply is required to inspect the initialized tool list.++- **Unreachable MCP degrades initialization (`F17.agent-uses-server`).** Beside the actual echo fixture arrange first a missing stdio executable, then a `streamable-http` server at `http://127.0.0.1:1/mcp`; for each, start a chat through Home with the dummy loopback model and inspect `conversation events <id> --kinds SystemPromptEvent --from-start --last 1`. Expected: initialization completes and retains `qa_settings_echo_qa_echo`, then reaches the refused LLM endpoint. This checks connection-failure degradation before a successful reply. Structural invalid config has the separate [OpenHands/software-agent-sdk#5643](https://github.com/OpenHands/software-agent-sdk/issues/5643) failure above. Remove only the own bad entry.++- **MCP timeout leaves initialization (`F17.agent-uses-server`).** An own stdio process that keeps stdin open for 90 seconds without answering the MCP handshake is a real unresponsive-server precondition. Configure it beside the echo fixture, then start a chat from Home using the dummy loopback model. Expected: after the bounded initialization timeout, a SystemPromptEvent and user message persist and the run reaches the refused LLM endpoint. Inspect the initialized tools separately: proceeding past the timeout alone does not establish healthy-sibling retention or a successful model reply. Remove the own slow entry and doctor.+## Gotchas-- Server errors on MCP mutations toast the server's message, or `An error occurred` when the body has none (#18051). Expected: one toast per failed action. The stale-editor delete and stale-card toggle 404 paths each produced one error toast on Local desktop and phone on 2026-10-12, after [#18182](https://github.com/OpenHands/OpenHands/pull/18182) fixed [#18181](https://github.com/OpenHands/OpenHands/issues/18181). Use the stale-mutation recipe above; this proof does not cover a backend-down 502 or the conversation Overview toggle. A custom server's save (**Save Server** in the edit form, **Add Server** in the add form) whose pre-save connection test fails because the Agent Server is down (`control-openhands service stop agent-server` with the editor open, then `control-openhands restart`) shows one `An error occurred` toast and keeps the editor open since [#18267](https://github.com/OpenHands/OpenHands/pull/18267) (closed #18266; verified 2026-10-13 on Light+ desktop); while the server is down, background `GET /api/settings` polls add their own `An error occurred` toasts every few seconds, so count only the toast that follows the click. A failed add, edit or delete is still sent three times (`withRetry` in `settings-service.api.ts`, 0.5 s and 1 s apart; a `POST 409` for a taken name shows three rows in `browser network`) before its single toast, the same `withRetry` root cause as open [#17932](https://github.com/OpenHands/OpenHands/issues/17932).+- Server errors on MCP mutations show one toast with the server's message, or `An error occurred` when the body has none. A custom server's **Save Server** or **Add Server** action whose pre-save connection test fails keeps the editor open. To check backend loss, stop the owned Agent Server with the editor open, click Save, then restore it with `control-openhands restart`; background settings polls can add separate error toasts, so count only the toast caused by the click. A failed add, edit or delete is retried three times (`withRetry` in `settings-service.api.ts`, 0.5 s and 1 s apart) before its single toast; that retry behavior is open [#17932](https://github.com/OpenHands/OpenHands/issues/17932).- Desktop `/customize` is a client-side redirect: `browser url` may still read `/customize` for a moment; use `wait-url '/mcp(\?|$)'`.- From 768 to 1023 px the sidebar rail stays and the phone top bar is gone. The compact Customize selector above each detail page provides sibling navigation; the rail's **Customize** link (`sidebar-skills-link`, unscoped at this width) still returns to the hub.- On a phone the drawer and the hidden desktop rail both render `sidebar-skills-link`; scope it with `testid=sidebar-mobile-drawer >> ...` or the click fails with a strict-mode error, even though `browser testids` lists it once.@@ -118,12 +126,15 @@ Preconditions:- `@modelcontextprotocol/server-everything sse` printed `Server is running` here but never answered on its port; use the `streamableHttp` mode for the remote-server fixture.- A server without a catalog credential probe reads `Reachable — credentials not verified`, never `Connected — credentials verified (N tool(s))`, even without credentials. Health after a save is seeded from the pre-save test and is lost on reload (`Not checked yet`).- Adding a custom server shows no toast; only catalog installs and deletes toast. Assert the card after reload.-- Catalog installs are add-only: a second Time install becomes `time_1`. Since #18034 (closed #17930) STDIO servers match the catalog by command plus leading args instead of by name, so `time_1` and a custom server that runs a catalog command under its own name (`qa_time`) keep that entry's description, badge and search keywords, while a server that only borrows a catalog name does not (a STDIO `fetch` running `qa-no-such-command` shows only its command) (`F17.install-stdio`, `F17.custom-add`). Twelve library entries (provider-OAuth-only, such as Sentry, HubSpot or the Cloudflare OAuth servers) open an empty modal whose Install does nothing (`F17.install-unsupported`) (#17922).-- The section filter's trigger and menu are named `Filter MCP servers` since #18035 (closed #17939); a checkout without it reuses the conversation filter's `Filter conversations`. The test ids work on both.-- Header and environment-variable format errors have distinct copy since [#17966](https://github.com/OpenHands/OpenHands/pull/17966) (closed [#17958](https://github.com/OpenHands/OpenHands/issues/17958)). `F17.custom-validation` expects `Headers must follow KEY=value format` for headers and `Environment variables must follow KEY=value format` for env lines. Verified 2026-10-11 on Local desktop and phone. Feedback uses the readable theme classes `text-feedback-error` and `text-feedback-success` since [#18270](https://github.com/OpenHands/OpenHands/pull/18270).+- Catalog installs are add-only: a second Time install becomes `time_1`. STDIO servers match the catalog by command plus leading args rather than name, so `time_1` and a custom server running a catalog command under its own name (`qa_time`) retain the description, badge and search keywords. A server that only borrows a catalog name does not (`F17.install-stdio`, `F17.custom-add`). Twelve provider-OAuth-only library entries, such as Sentry, HubSpot and Cloudflare OAuth, open an empty modal whose Install does nothing (`F17.install-unsupported`, [#17922](https://github.com/OpenHands/OpenHands/issues/17922)).+- The section filter's trigger and menu are named `Filter MCP servers`.+- Header and environment-variable format errors have distinct copy. `F17.custom-validation` expects `Headers must follow KEY=value format` for headers and `Environment variables must follow KEY=value format` for env lines. Feedback uses readable theme classes `text-feedback-error` and `text-feedback-success`.- A catalog stdio server runs with the Agent Server's minimal environment: behind a TLS-intercepting proxy its own HTTPS calls fail (`fetch failed`) even when npx could download it, and the install modal has no field for `NODE_EXTRA_CA_CERTS`; use the custom editor with the same command to see the provider's real answer (`F17.credential-probe`). The `Credential check failed: fetch failed` wording itself is #18161: the Slack interpreter turns any `is_error` tool result into a credential failure, though the server connected and listed its tools.-- GitHub's hosted MCP endpoint answers an invalid token with HTTP 401 at connect, before tools are listed, so its `get_me` probe never runs and the modal shows the connection wording (`Connection failed: HTTPStatusError … 401 Unauthorized`), not `Credential check failed`; filed as OpenHands/software-agent-sdk#5607, since the test endpoint reports the 401 as `error_kind` `unknown`.-- The shell-gated `qa_vault` fixture exits before the MCP handshake, which the Agent Server reports either at once as `Connection failed: McpError: Connection closed` or only at its deadline as `Connection timed out. Check the URL and try again.` (the timed-out wording was seen once on 2026-10-07; on 2026-10-08 all 21 tries answered `Connection closed` within 0.4 s); neither is the `Could not start the server command` wording, which needs a command that cannot be spawned at all. Wait on `text=Connection`, which matches both and not `Connected — …`.+- GitHub's hosted MCP endpoint answers an invalid token with HTTP 401 at connect, before tools are listed, so its `get_me` probe never runs and the modal shows the connection wording (`Connection failed: HTTPStatusError … 401 Unauthorized`), not `Credential check failed`; filed as [OpenHands/software-agent-sdk#5607](https://github.com/OpenHands/software-agent-sdk/issues/5607), since the test endpoint reports the 401 as `error_kind` `unknown`.+- The shell-gated `qa_vault` fixture exits before the MCP handshake, which the Agent Server reports either at once as `Connection failed: McpError: Connection closed` or only at its deadline as `Connection timed out. Check the URL and try again.`; neither is the `Could not start the server command` wording, which needs a command that cannot be spawned at all. Wait on `text=Connection`, which matches both and not `Connected — …`.- `browser network --bodies` shows what the page sent, never a secret: a value under a key that names a credential (key, token, secret, auth, pass, session, sig, credential) and every value of an `env` or `headers` map becomes `<redacted N chars>` (N the real length), and a URL in any other string (a server's `url`) keeps its host and path but loses its userinfo and the values of its secret-named query parameters (`https://<redacted 2 chars>:<redacted 8 chars>@host/mcp?api_key=<redacted 17 chars>`), while the settings API's `**********` placeholder is kept as is, so a row tells "placeholder sent" from "a value sent". The MCP editor sends neither for an untouched secret: Test connection swaps the placeholder for the stored encrypted value (a `GET /api/settings` in `encrypted` mode first; Fernet tokens, so 120 chars for a 19-char token, 140 for 40, 268 for 142) and Save leaves `env` out of its sparse PATCH altogether (`F17.stored-secret-reuse`). Without `--bodies` no row carries a `body` field, and only app-origin writes keep one.- A browser verb's client call waits 120 s, or 10 s longer than the verb's own `--timeout` or `--observe-ms` when that is longer, so `browser wait 'testid=mcp-install-modal-error' --timeout 150000` runs its full deadline and a miss ends with Playwright's timeout and a failure screenshot (before this, every call was aborted at 120 s with `TimeoutError: The operation was aborted due to timeout` while the page kept going). For a first Slack install use a plain click, then that `wait`; `--observe-ms` is for a few transient states, not a minute-long outcome.- Known issue #17941: turning off an enabled card toggle can drop the click when the pointer moves in and presses at once (`F17.enable-disable`).+- Known issue [#15421](https://github.com/OpenHands/OpenHands/issues/15421): A configured Linear MCP can be ignored while an explicitly disabled Linear skill is used instead. The new-chat and automation paths need a working authorized Linear MCP connection, a concrete issue ID and an approved model; the report's automation ZIP is also needed to replay its exact automation settings. (`F17.agent-uses-server`).+- Known issue [#15430](https://github.com/OpenHands/OpenHands/issues/15430): An OAuth MCP callback in Docker can redirect to an unexposed dynamic localhost port. Completing the callback needs a Docker stack and an approved OAuth MCP provider/account (the report uses GitLab); a local unauthenticated Test connection cannot reach the redirect. (`F17.install-oauth`).+- Known issue [OpenHands/software-agent-sdk#5375](https://github.com/OpenHands/software-agent-sdk/issues/5375): A successful GitLab OAuth login can be followed by an invalid localhost redirect URI on the first ticket-to-code-change MCP action. Reproduction needs the authorized self-hosted GitLab MCP endpoint/account and the ticket-to-code-change plugin/workflow; the reported endpoint is not a public fixture. (`F17.install-oauth`, `F17.agent-uses-server`).diff --git a/.agents/skills/verify-openhands/references/feature-map/F18-skills.md b/.agents/skills/verify-openhands/references/feature-map/F18-skills.mdindex 3cfdb79c66..f99e2487c1 100644--- a/.agents/skills/verify-openhands/references/feature-map/F18-skills.md+++ b/.agents/skills/verify-openhands/references/feature-map/F18-skills.md@@ -87,6 +87,14 @@ Preconditions:- **Cloud link (`F18.cloud-link`).** Blocked: needs a Cloud backend. Expected: `control-openhands browser attr 'testid=sidebar-extensions-/skills' href` is `<cloud host>/settings/skills` with `target` `_blank`, and the Plugins and Apps items are hidden.- **Restore.** Leave `add-javadoc` and `flarglebargle` off and `qa-hello` on; the fixture skill and the conversations vanish with the run's private state.+- **Colliding skill names do not leave ghost cards (`F18.card`, `F18.facets`).** Arrange `fixture skill --name skill-creator --trigger qa-collision-ping` in the run's isolated personal folder beside the bundled same-name skill, plus `qa-toggle`. Reload `/skills`, search `skill-creator`, then search `qa-toggle`; repeat the two searches. Expected: only the one matching toggle card and no `skill-card-skill-creator`. Known failure, reproduced 2026-10-12 ([#18287](https://github.com/OpenHands/OpenHands/issues/18287)): `skills-result-summary` says `1 result(s)` while two then three stale same-source skill-creator cards remain. Read their source titles and screenshot the mismatched list. Remove only the own colliding fixture and reload to clear ghosts.++- **Skill state survives a delayed sample (`F18.toggle`).** With an own enabled `qa-toggle` fixture, use its real switch with `--hover-first 250` when removing; drive disabled → enabled → disabled, observe for eight seconds and read `aria-checked`, settings `disabled_skills` and a reload. Expected: the switch stays false and the server keeps `qa-toggle` disabled across reload. Repeat the cycle, then restore enabled; a bounded sample does not establish every possible delay.++- **Backend switch refreshes skill data (`F18.page`).** Use Local and the disposable `QA_SCOPE` identity of the own stack described in F17. Load/reload the Local Skills page once, then arrange a new personal fixture `qa-scope-new-3`; clear the skills network log. Switch to QA_SCOPE through the actual selector and inspect the still-mounted Skills page, with no `goto` or reload between the switch and observation. Expected: a skills request for the selected backend and the new card. Known failure, reproduced twice 2026-10-12 ([#16893](https://github.com/OpenHands/OpenHands/issues/16893)): zero `/api/skills` requests and card count 0; a subsequent reload requests actual data and shows the card (1). Repeat in the other direction with a new marker `qa-scope-new-4`. Two registry identities / real fixture changes share one physical run here; separate Cloud data and conversation-scoped caches are further branches. Restore Local and remove only the fixtures/alias.++- **Persistence folder skill loads (`F18.personal-skill-dirs`).** Arrange a valid `qa-legacy` fixture under `private/skills/qa-legacy/SKILL.md` (move the CLI-created personal fixture there; `private` is this run's persistence directory). Reload Skills, read the card/source path and start a chat through Home with `qa-legacy-ping. Do not run tools; reply in one line.` using the dummy loopback profile. Expected: the card appears and the initialized SystemPromptEvent/user message includes `qa-legacy` in available/activated skills. This checks discovery and activation before the model call. Remove the own skill afterwards.+## Gotchas- `sidebar-skills-link` is labelled **Customize** and lands on `/mcp` on desktop (`/customize` redirects); Skills is one more click. Test ids embed the route: quote `'testid=sidebar-extensions-/skills'`.@@ -100,10 +108,11 @@ Preconditions:- Skill lists are cached for 10 minutes: `browser reload` (or `goto`) after the agent creates a skill file, or the new card does not appear.- `navigator.clipboard.readText()` through `browser eval` hangs (no clipboard permission in the page); read what a copy button wrote with `control-openhands browser clipboard` and assert the button's `aria-label` too. The app runs on `127.0.0.1` (a secure context), so the unguarded `clipboard.writeText` calls cannot be tested for insecure-origin failures here.- With the Agent Server stopped, reloading `/skills` shows the "Manage backends / Disconnected" screen instead of the page: load the page first, then stop the service. After the failed save the card keeps showing the new state until a reload (#17941).-- Before #18035 (closed #17939) the page description said "Search from the sidebar to filter the list" and offered curl and install flows the page does not have; a checkout without it still does.-- Escape from the modal's focused `+N` overflow trigger closes only the popover and restores trigger focus; the next Escape closes the detail modal. Verified 2026-10-11 on Local desktop and phone after [#17967](https://github.com/OpenHands/OpenHands/pull/17967) (closed [#17957](https://github.com/OpenHands/OpenHands/issues/17957)). The card popover uses the same Escape behavior.-- The Add skill example, its Copy button and the add-skill **Use skill** message share one constant. Since #18035 it is `.../skills/code-review`; before, it pointed at `skills/codereview`, which is 404 upstream (the skill was renamed).+- Escape from the modal's focused `+N` overflow trigger closes only the popover and restores trigger focus; the next Escape closes the detail modal. The card popover uses the same Escape behavior.+- The Add skill example, its Copy button and the add-skill **Use skill** message share the `.../skills/code-review` path.- Skill changes apply to new conversations only; an open conversation keeps the skills it started with.- `conversation events --grep` searches whole events, the user's message text included: a trigger that contains the skill name (`qa-gone-ping` for `qa-gone`) matches the user `MessageEvent` even when the skill is not loaded. Give fixture skills a trigger that does not contain their name, or grep for `<name>qa-gone</name>`.- `fixture skill --repo` writes the project `SKILL.md` and records it in the repo only with `--commit` (`committed.sha`; a re-run reports `unchanged`); a New Worktree conversation checks out the repo's commits, so pass `--commit`. Workspace skills are loaded per conversation and never appear on `/skills`.- Without an LLM key the skill set is still observable: the `SystemPromptEvent` (available skills), the user `MessageEvent` (`activated_skills`) and the chat's **Skill Ready** row (`generic-event-message-title`; Expand lists the skills under `Triggered Skill Knowledge:`) all exist before the first model call, after which the conversation ends in `error`.+- Known issue [#16298](https://github.com/OpenHands/OpenHands/issues/16298): A Slack-monitor setup can leave the agent without the expected skill. The report omits the concrete monitor setup and loaded skill name/version; reproducing the runtime also needs an approved model and a configured Slack integration. Listing the bundled catalog alone cannot settle it. (`F18.skill-in-chat`).+- Known issue [#15458](https://github.com/OpenHands/OpenHands/issues/15458): The global Skills page can list AGENTS.md files from cloned automation-run worktrees. The report says the discovery trigger is unknown; replay needs the affected automation workspace/discovery-root layout or captured request parameters, rather than only a normal project-skill fixture. (`F18.page`, `F18.project-skill`).diff --git a/.agents/skills/verify-openhands/references/feature-map/F19-plugins.md b/.agents/skills/verify-openhands/references/feature-map/F19-plugins.mdindex d42026f99e..fbd8a6c364 100644--- a/.agents/skills/verify-openhands/references/feature-map/F19-plugins.md+++ b/.agents/skills/verify-openhands/references/feature-map/F19-plugins.md@@ -68,7 +68,7 @@ Preconditions:- **Install from a card (`F19.install-card`).** Run `control-openhands browser toasts --clear`, `control-openhands browser click 'testid=plugin-install-magic-test'`, `control-openhands browser wait 'testid=plugin-toggle-magic-test' --timeout 60000` and `control-openhands browser toasts` (`Plugin installed.`). `control-openhands browser text 'testid=plugin-version-magic-test'` is `v1.0.0` and `control-openhands browser eval "[...document.querySelectorAll('[data-testid^=plugin-card-]')].map(e=>e.dataset.testid)[0]"` is `plugin-card-magic-test`. After `control-openhands browser reload` and `control-openhands browser click 'testid=plugins-filter-installed'`, `browser count '[data-testid^="plugin-card-"]'` is `1`, and `control-openhands api GET /api/plugins/installed` lists `magic-test` with `"enabled": true`. Click `testid=plugins-filter-all`.- **Enable switch (`F19.enable-toggle`).** Run `control-openhands browser attr 'testid=plugin-toggle-magic-test' aria-checked` (`true`) and `control-openhands browser attr 'testid=plugin-toggle-magic-test' aria-label` (`Disable plugin`), `control-openhands browser hover 'testid=plugin-toggle-magic-test'`, `control-openhands browser click 'testid=plugin-toggle-magic-test'`, wait three seconds, then `browser attr ... aria-checked` again (`false`, with `aria-label` `Enable plugin`); after `control-openhands browser reload` it is still `false` and `api GET /api/plugins/installed` shows `"enabled": false`. Today the disable is often lost: the reload's concurrent reads hit the metadata race (Gotchas) and the API, then the switch, read `"enabled": true` again (2 of 3 trials; the PATCH itself returns 200). If that happens, disable it once more before the modal step, which needs the plugin disabled. In the modal: `control-openhands browser click 'testid=plugin-card-magic-test'`, `control-openhands browser click 'testid=plugin-detail-modal >> text="Disabled"'`, then `control-openhands browser eval "document.querySelector('[data-testid=plugin-modal-toggle-magic-test]').checked"` (`true`); the API shows `"enabled": true` and, after closing the modal, the card switch's `aria-checked` is `true`.- **Enabled plugins auto-load (`F19.enabled-autoload`).** With `magic-test` enabled, run `control-openhands conversation start --prompt "alakazam" --wait --timeout 240` and `control-openhands browser text 'testid=agent-message >> nth=-1'`: `The magic worked! Plugin loaded successfully. 🎩✨`. Then `control-openhands browser goto /plugins`, disable it with the card switch (hover, click; API `"enabled": false`) and run the same `conversation start`: the reply is an ordinary request for a task, without the phrase. `browser goto /plugins` again; `magic-test` stays installed and disabled. Today the revisit often re-enables it (API `"enabled": true`, `installed_at` reset; see Gotchas), so re-check the API before relying on the disabled state.-- **Update (`F19.refresh`).** Run `control-openhands browser click 'testid=plugin-card-magic-test'`, `control-openhands browser toasts --clear`, `control-openhands browser click 'testid=plugin-detail-refresh-magic-test'`, wait a few seconds, then `control-openhands browser toasts`. Expected `Plugin updated.` (the toast fades fast: read `control-openhands browser toasts --history`). Whether it works depends on the recorded source (`control-openhands browser text 'testid=plugin-source-magic-test'`): with the catalog path it does. Known failure (reproduced 2026-10-09): a page revisit after a catalog install usually rewrites the source to `local` (OpenHands/software-agent-sdk#5496, see Gotchas), and Update then answers 500 with one toast carrying the server's reason, `Unable to parse extension source: local. Expected formats: 'github:owner/repo', git URL, or local path` (`browser screenshot --feature F19.refresh --name local-source-error`). On a plugin whose source survived (the GitHub install in **Add plugin from GitHub**, `browser text 'testid=plugin-source-magic-test'` = `github:OpenHands/extensions`), the same click logs `Updating extension magic-test from github:OpenHands/extensions` and `toasts --history` shows `Plugin updated.` Close the modal.+- **Update (`F19.refresh`).** Run `control-openhands browser click 'testid=plugin-card-magic-test'`, `control-openhands browser toasts --clear`, `control-openhands browser click 'testid=plugin-detail-refresh-magic-test'`, wait a few seconds, then `control-openhands browser toasts`. Expected `Plugin updated.` (the toast fades fast: read `control-openhands browser toasts --history`). Whether it works depends on the recorded source (`control-openhands browser text 'testid=plugin-source-magic-test'`): with the catalog path it does. Known failure (reproduced 2026-10-09): a page revisit after a catalog install usually rewrites the source to `local` ([OpenHands/software-agent-sdk#5496](https://github.com/OpenHands/software-agent-sdk/issues/5496), see Gotchas), and Update then answers 500 with one toast carrying the server's reason, `Unable to parse extension source: local. Expected formats: 'github:owner/repo', git URL, or local path` (`browser screenshot --feature F19.refresh --name local-source-error`). On a plugin whose source survived (the GitHub install in **Add plugin from GitHub**, `browser text 'testid=plugin-source-magic-test'` = `github:OpenHands/extensions`), the same click logs `Updating extension magic-test from github:OpenHands/extensions` and `toasts --history` shows `Plugin updated.` Close the modal.- **Start a conversation from an installed plugin (`F19.start-conversation`).** Run `control-openhands browser click 'testid=plugin-card-magic-test'`, `control-openhands browser click 'testid=plugin-detail-start-conversation-magic-test' --expect-url '/launch\?plugins='` and `control-openhands browser eval "JSON.parse(atob(new URLSearchParams(location.search).get('plugins')))"`. Expected: the plugin's real source. Today it is `[{"source":"local","ref":null,"repo_path":null}]` and the modal title is `Launch local`; after `control-openhands browser check 'testid=trust-checkbox'` and `control-openhands browser click 'testid=start-conversation-button' --expect-url '/conversations/[0-9a-f-]+'`, `control-openhands browser type 'testid=chat-input' 'alakazam'` and `control-openhands browser press Enter --selector 'testid=chat-input'` leave the message at `Sending...` (still after 25 s) with the banner `Unable to parse plugin source: local. Expected formats: 'github:owner/repo', git URL, or local path` (`control-openhands browser screenshot --feature F19.start-conversation --name local-source-conversation`). When the installed plugin kept its source (the GitHub install below), the link carries it: `[{"source":"github:OpenHands/extensions","ref":"<resolved commit>","repo_path":"plugins/magic-test"}]` and the title is `Launch magic-test`.- **Uninstall (`F19.uninstall`).** Run `control-openhands browser goto /plugins`, `control-openhands browser toasts --clear`, `control-openhands browser click 'testid=plugin-card-magic-test'`, `control-openhands browser click 'testid=plugin-detail-uninstall-magic-test'`, then `control-openhands browser count 'testid=plugin-detail-modal'` (`0`), `control-openhands browser count 'testid=confirmation-modal'` (`0`: no confirmation), `control-openhands browser toasts` (`Plugin uninstalled.`) and `control-openhands browser count 'testid=plugin-install-magic-test'` (`1`). After `control-openhands browser reload` the Install button is still there, `browser text 'testid=plugin-source-magic-test'` is the catalog path again and `api GET /api/plugins/installed` returns `{"plugins": []}`.- **Install from the modal (`F19.install-modal`).** Run `control-openhands browser toasts --clear`, `control-openhands browser click 'testid=plugin-card-city-weather'`, `control-openhands browser click 'testid=plugin-detail-install-city-weather'`, `control-openhands browser wait 'testid=plugin-detail-uninstall-city-weather' --timeout 60000` and `control-openhands browser toasts` (`Plugin installed.`). The modal stays open (`browser count 'testid=plugin-detail-modal'` is `1`), `control-openhands browser eval "document.querySelector('[data-testid=plugin-modal-toggle-city-weather]').checked"` is `true`, and `control-openhands browser snapshot 'testid=plugin-detail-modal >> footer'` lists **Update**, **Uninstall**, **Start Conversation**, **Close**. Clean up with `control-openhands browser click 'testid=plugin-detail-uninstall-city-weather'`; `browser count 'testid=plugin-install-city-weather'` is `1`.@@ -87,10 +87,43 @@ Preconditions:- **Phone (`F19.phone`).** Run `control-openhands browser viewport phone`, `control-openhands browser goto /plugins`, `control-openhands browser bbox 'testid=skills-plugins-screen'` (`insideViewport` `true`, `pageHorizontalOverflow` `false`) and `control-openhands browser eval "(()=>{const s=document.querySelector('[data-testid=plugins-search-input]').getBoundingClientRect(), f=document.querySelector('[data-testid=plugins-status-filter]').getBoundingClientRect();return {searchBottom:Math.round(s.bottom), filterTop:Math.round(f.top), filterWidth:Math.round(f.width)}})()"`: `filterTop` is below `searchBottom` (stacked) and `filterWidth` is `348`. `control-openhands browser screenshot --feature F19.phone --name list` shows one column of cards. Open `testid=plugin-card-city-weather` and `control-openhands browser bbox 'testid=plugin-detail-modal'` (`insideViewport` `true`, `browser screenshot --feature F19.phone --name detail`), Escape; `control-openhands browser click 'testid=plugins-add-plugin-button'` and `control-openhands browser bbox 'testid=add-plugin-modal'` (`insideViewport` `true`), Escape. The parameters link from above gives `control-openhands browser bbox 'testid=plugin-launch-modal'` `insideViewport` `true` (`browser screenshot --feature F19.phone --name launch`). Hub entry: `control-openhands browser goto /customize`, `control-openhands browser click 'testid=extensions-mobile-hub >> testid=sidebar-extensions-/plugins' --expect-url '/plugins(\?|$)'` and `control-openhands browser click 'testid=sidebar-mobile-back-button' --expect-url '/customize(\?|$)'`. Return with `control-openhands browser viewport desktop`.- **Errors sweep (`F19.page`).** `control-openhands browser errors --app-only` shows no page errors; the only app HTTP errors are the deliberate ones (refresh 500, install 400, 502s while the Agent Server was stopped).+- **Launch trust shows actual host (`F19.launch-trust`).** Run `control-openhands browser goto '/launch?plugin_source=https%3A%2F%2Fevil.example%2Fgithub.com%2FAll-Hands-AI%2Fopenhands-plugin'` and `browser text 'testid=plugin-launch-modal'`. Expected: the trust sentence contains the full `https://evil.example/...` source. Known failure, reproduced 2026-10-12 ([#17162](https://github.com/OpenHands/OpenHands/issues/17162)): it shows `All-Hands-AI/openhands-plugin` and omits `evil.example` throughout the dialog. `browser enabled 'testid=start-conversation-button'` is `false`. Screenshot, reload and repeat the modal text; leave without ticking trust or starting. This display check does not download or run that source.++- **Unicode parameter round-trip (`F19.launch-unicode`).** Open `/launch?plugins=W3sic291cmNlIjogImh0dHBzOi8vZ2l0aHViLmNvbS9leGFtcGxlL3BsdWdpbiIsICJwYXJhbWV0ZXJzIjogeyJ0YXNrIjogIuaVtOeQhuWPkeW4g%2BivtOaYjiDwn5qAIn19XQ%3D%3D` (UTF-8 base64 of `[{"source":"https://github.com/example/plugin","parameters":{"task":"整理发布说明 🚀"}}]`). Run `browser value 'testid=plugin-0-param-task'`, screenshot and reload/read again. Expected: exactly `整理发布说明 🚀`. Known failure, reproduced 2026-10-12 ([#16735](https://github.com/OpenHands/OpenHands/issues/16735)): both reads contain Latin-1 mojibake. Leave without trust or Start. This tests the live launch decoder; the source-side URL builder's exception is a separate branch.++- **Cached source rejects an unknown ref (`F19.add-plugin`).** Arrange a run-owned local Git plugin with `.plugin/plugin.json` name `qa-pin-plugin`, a valid skill, tag `v1` at version 1.0.0 and a later main commit at version 2.0.0. Add its `file://` source with Ref `v1` through the real form and uninstall it, leaving the Git cache warm. Add the same source with Ref `qa-does-not-exist`; inspect the modal, toast and installed version, then repeat after uninstalling. Expected: missing-ref error and no installation. Known failure, reproduced twice 2026-10-12 ([OpenHands/software-agent-sdk#5645](https://github.com/OpenHands/software-agent-sdk/issues/5645)): the modal closes with `Plugin installed.` and cached version 1.0.0 is installed. No external Git host or account is required.++- **Update retains a requested pin (`F19.refresh`).** Using that own Git fixture, install Ref `v1` and privately read only its public `requested_ref` metadata in `private/plugins/installed/.installed.json`; it is `v1`. Click the card's Update and wait for the completed `Plugin updated.` toast before reading again. Expected: retain `v1`. Known failure, reproduced 2026-10-12 ([OpenHands/software-agent-sdk#4363](https://github.com/OpenHands/software-agent-sdk/issues/4363)): the stored requested ref becomes null. The visible version remained 1.0.0 in this cache, so this diagnostic proves loss of the requested pin, not a default-HEAD version jump. Earlier completed updates also had null; keep that limit explicit and uninstall the fixture.++- **A malformed optional plugin does not block the first message (`F19.enabled-autoload`).** Arrange a local own plugin directory with an invalid JSON `.plugin/plugin.json` and open its `plugin_source` launch path. Review/trust only that dummy local fixture and Start; creation yields an idle conversation because plugin loading is deferred. Send a real first message with the dummy loopback model, then read the UI error card, `conversation events <id> --kinds SystemPromptEvent --from-start --last 1` and status; repeat with a fresh launch. Expected: skip/report the broken optional plugin and continue initialization. Known failure, reproduced twice 2026-10-12 ([OpenHands/software-agent-sdk#5718](https://github.com/OpenHands/software-agent-sdk/issues/5718)): the UI shows `Invalid JSON` and `Sending...`, no event/system prompt is persisted and status stays idle. A valid manifest with malformed SKILL.md instead did initialize and reached the refused model endpoint; do not merge those different outcomes. The fixture contains no runnable entry point or external dependency; remove it afterwards.++- **Disabled plugin reconciles on resume (`F19.enabled-autoload`).** Arrange an own installed `qa-resume-plugin` with a valid `.plugin/plugin.json`, `commands/qa-hello.md`, `skills/qa-plugin-skill/SKILL.md` and a `UserPromptSubmit` hook containing only `echo QA_HOOK`. Start a real Home chat with the dummy loopback profile; inspect its HookExecutionEvent (one) and plugin skill names through `api GET '/api/conversations/<id>?include_skills=true' --pick agent.agent_context.skills` (keep the full response private). Disable the real `plugin-toggle-qa-resume-plugin` with `--hover-first 250`, verify `GET /api/plugins/installed` says false, leave Plugins, `restart`, `doctor`, reload the conversation and `conversation send <id> --prompt '/qa-resume-plugin:qa-hello'`. Expected: the disabled command does not activate and plugin skills disappear. Known failure, reproduced in two independent persisted conversations on 2026-10-12 ([OpenHands/software-agent-sdk#5628](https://github.com/OpenHands/software-agent-sdk/issues/5628)): the resumed user MessageEvent activated `qa-resume-plugin:qa-hello`, and both plugin skills remained; the original hook count stayed one, so hooks did follow disable. A fresh chat begun while disabled activated no command and ran no hook. These are real pre-model activation/state observations; no successful reply or plugin MCP reconciliation is claimed. Uninstall only the own plugin and remove its source fixture afterwards.++- **Unavailable plugin source is a handled Update error (`F19.refresh`).** Expected:400 for a vanished source, with the installation unchanged. Known failure (reproduced twice 2026-10-16, [OpenHands/software-agent-sdk#5897](https://github.com/OpenHands/software-agent-sdk/issues/5897)): two real Update clicks after moving only the owned source aside showed the missing-path error and returned500; installed version/enabled state stayed unchanged. This is the plugin branch; Canvas has no skill-management Refresh action.++ ```sh known-bug F19.refresh missing-source+ QA_PLUGIN_SOURCE="$OH_VERIFY_RUN/workspace/qa-known-refresh-plugin"+ mkdir -p "$QA_PLUGIN_SOURCE/.plugin"+ printf '%s\n' '{"name":"qa-known-refresh-plugin","version":"0.1.0","description":"Inert disposable Update check."}' > "$QA_PLUGIN_SOURCE/.plugin/plugin.json"+ trap 'if [ -d "$QA_PLUGIN_SOURCE.hidden" ]; then mv "$QA_PLUGIN_SOURCE.hidden" "$QA_PLUGIN_SOURCE"; fi; control-openhands api DELETE /api/plugins/installed/qa-known-refresh-plugin --write >/dev/null; rm -rf "$QA_PLUGIN_SOURCE"; control-openhands browser reload >/dev/null' EXIT+ control-openhands browser viewport desktop+ control-openhands browser goto /plugins+ control-openhands browser click 'testid=plugins-add-plugin-button'+ control-openhands browser fill 'testid=add-plugin-source-input' "$QA_PLUGIN_SOURCE"+ control-openhands browser click 'testid=add-plugin-submit'+ control-openhands browser wait 'testid=add-plugin-modal' --state hidden --timeout 30000+ control-openhands assert --field status --equals 200 -- api GET /api/plugins/installed/qa-known-refresh-plugin+ control-openhands browser click 'testid=plugin-card-qa-known-refresh-plugin'+ mv "$QA_PLUGIN_SOURCE" "$QA_PLUGIN_SOURCE.hidden"+ control-openhands browser network --clear+ control-openhands browser click 'testid=plugin-detail-refresh-qa-known-refresh-plugin' --observe 'role=status' --observe-ms 1000+ control-openhands assert --field recent.0.status --equals 400 --bug 'recent.0.status eq 500 && recent.0.path eq "/api/plugins/installed/qa-known-refresh-plugin/refresh"' -- browser network --filter '/api/plugins/installed/qa-known-refresh-plugin/refresh' # bug+ ```+## Gotchas- The card switch swaps its icon on hover. A plain `browser click` that moves the pointer onto it from elsewhere (for example right after clicking a filter segment) is often swallowed (#17941): no `PATCH /api/plugins/installed/<name>` shows in `browser network`. Run `browser hover` on the switch first, then click, and assert `aria-checked` and the API, not the click result.-- Installed plugins lose their source and their enabled state, intermittently (OpenHands/software-agent-sdk#5496). The Agent Server rewrites `<run>/private/plugins/installed/.installed.json` with a plain truncate-and-write on every load (SDK `InstallationMetadata.save_to_dir`), so a concurrent `GET /api/plugins/installed` (page loads fire several) can read an empty file: it logs `Failed to load installed extension metadata … Invalid JSON: EOF while parsing` and `Discovered untracked extension`, then saves every plugin with `"source": "local"`, `"enabled": true` and a fresh `installed_at`. It hit catalog installs in most trials and a disabled plugin on 2 of 3 page revisits; GitHub installs kept their source in this run. Once it hits, the card shows `local`, **Update** fails with a 500, **Start Conversation** builds an unusable `/launch` link, and a disabled plugin silently auto-loads again. Prove the launch path from a card that is not installed, and re-read the API after every navigation.+- Installed plugins lose their source and their enabled state, intermittently ([OpenHands/software-agent-sdk#5496](https://github.com/OpenHands/software-agent-sdk/issues/5496)). The Agent Server rewrites `<run>/private/plugins/installed/.installed.json` with a plain truncate-and-write on every load (SDK `InstallationMetadata.save_to_dir`), so a concurrent `GET /api/plugins/installed` (page loads fire several) can read an empty file: it logs `Failed to load installed extension metadata … Invalid JSON: EOF while parsing` and `Discovered untracked extension`, then saves every plugin with `"source": "local"`, `"enabled": true` and a fresh `installed_at`. It hit catalog installs in most trials and a disabled plugin on 2 of 3 page revisits; GitHub installs kept their source in this run. Once it hits, the card shows `local`, **Update** fails with a 500, **Start Conversation** builds an unusable `/launch` link, and a disabled plugin silently auto-loads again. Prove the launch path from a card that is not installed, and re-read the API after every navigation.- A failed plugin install, update or uninstall shows one toast with the server's reason (enable uses the same handler; its failure is not driven here). A response with no reason in it, such as the ingress's 502 while the Agent Server is stopped, shows the generic `An error occurred`; read the cause in the run's `private/stack.log`.- Installing a plugin whose name is already installed points the user to **Update** or **Uninstall**, as Apps do (`F20.install-duplicate`): with the catalog `magic-test` installed, **Add plugin** with `github:OpenHands/extensions` and Path `plugins/magic-test` answers 409 (`browser network --filter 'plugins/install'`) and `browser toasts --history` holds one toast, `This plugin is already installed. Open its card to Update or Uninstall it.`, instead of the server's `force=true` hint; the modal stays open with the typed values (`browser screenshot --feature F19.add-plugin-error --name duplicate-409`). Other install errors still toast the server's reason (`F19.add-plugin-error`). The 409 adds one `http-error` and one `console.error` to `browser errors --app-only`.- **Uninstall** acts immediately, with no confirmation (MCP server delete and App uninstall confirm).@@ -100,3 +133,4 @@ Preconditions:- Build deep links with URL-encoded base64 (`%2B`, `%2F`, `%3D`): a raw `+` decodes to a space and gives the invalid-format error. Non-ASCII text must not be used in payloads (`atob` decodes bytes as Latin-1).- The page loads three queries (marketplace, installed, local) behind skeleton rows; wait for `testid=plugins-toolbar` before counting cards. With no plugins at all the page shows `No plugins found.` (`plugins-empty`), which the bundled catalog makes unreachable; on a Cloud backend **Add plugin** and card actions are disabled (not driven: no Cloud account).- Local plugins under `~/.agents/plugins` appear on the next page load without a restart; the Agent Server's `HOME` is `<run>/private/home`.+- Known issue [OpenHands/software-agent-sdk#5985](https://github.com/OpenHands/software-agent-sdk/issues/5985): Update can re-enable a disabled plugin when an installed-list request overlaps the force-reinstall directory swap. A live check needs an owned disabled plugin and a real concurrent list that lands in that window; the report's monkeypatched `copytree` schedule is not Canvas live proof. No overlap has been proved in this pass. (`F19.refresh`, `F19.enable-toggle`).diff --git a/.agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md b/.agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.mdindex aa5bfc173e..c62d12cfe6 100644--- a/.agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md+++ b/.agents/skills/verify-openhands/references/feature-map/F20-canvas-apps.md@@ -81,11 +81,14 @@ Preconditions:- **List error with Retry (`F20.load-error`).** Blocked: it needs an Agent Server whose `GET /api/canvas-extensions/installed` fails with a non-404 while its health checks pass. `control-openhands service stop agent-server` then `control-openhands browser goto /apps` shows the app-wide **Manage backends** gate (`Disconnected`) instead of the page, and on an already loaded page the cached list stays on screen.- **Unsupported backends (`F20.unsupported`).** Blocked: needs a Cloud backend account (`Apps are not available on Cloud backends yet.`, Apps hidden from Customize), a state with no backend (`Add an Agent Server backend to use Apps.`), or an Agent Server that passes Canvas's version gate (1.47.0 or newer) but has no `/api/canvas-extensions` routes (`This Agent Server does not support Apps yet. Upgrade the backend and try again.`). In each case **Add app** should be disabled. Released servers cannot produce the last state: the routes ship from 1.44.1, and an older server such as `control-openhands launch --new --sdk-ref v1.43.1` (whose `GET /api/canvas-extensions/installed` is `404`) never reaches `/apps`, because every page shows the **Manage backends** gate with `Agent Canvas requires agent-server 1.47.0 or newer; this backend is running 1.43.1.` (`doctor` still reports `ok`).+- **Malformed tree-URL escapes (`F20.git-tree-url`, `F20.install-error`).** On `/apps`, open `canvas-extensions-add-button`, fill `add-canvas-extension-source-input` with `https://github.com/o/r/tree/main%/apps/demo`, and click `add-canvas-extension-submit`. Inspect `browser errors --app-only`, `browser toasts --history`, and the modal count. Repeat with `https://gitlab.com/group/repo/-/tree/main/apps/%E0%A4%A`. Expected: submission produces a usable install/validation outcome. Known failure, reproduced 2026-10-12 on `fa1f724` with Agent Server 1.54.0: each submit raises `URI malformed`, emits no toast, skips the install request, and leaves the modal open ([#17783](https://github.com/OpenHands/OpenHands/issues/17783)). A reviewed local demo source still installs successfully as the control; no malformed source is executed.+- **Extension page home pin (`F20.rail-entry`).** Install and enable the reviewed local `src/fixtures/canvas-extensions/demo-page` through the form and trust confirmation. Hover `sidebar-canvas-extension-demo-page-hello`, inspect `browser testids 'role=complementary' --filter pin`, and compare with hovering `sidebar-automations-link`. Expected: the enabled extension page offers **Pin as home page** and the normal home-pin lifecycle. Known failure, reproduced 2026-10-12 on `fa1f724`: only Automate exposes `sidebar-pin-home-toggle-automations`; the extension page has no pin control ([#16995](https://github.com/OpenHands/OpenHands/issues/16995)).+## Gotchas- The card's refresh button (`canvas-extension-refresh-<name>`) is labelled **Update**, not "Refresh".-- The trust confirmation has no title and reuses the shared `confirmation-modal`; its body repeats the amber notice. Keyboard focus is not moved into it: after Space on the switch, Tab lands on the card's **Update** button behind the dialog.-- `nav_label` is dropped by the Agent Server: `GET /api/canvas-extensions/installed` returns pages with only `id`, `title` and `path` (Agent Server 1.50.1), so the rail shows the page title (OpenHands/software-agent-sdk#5501). Keep `F20.rail-label` failing until the backend returns it.+- The trust confirmation has no title and reuses the shared `confirmation-modal`; its body repeats the amber notice. Space on the disabled app's Enable switch focuses **Cancel**. Tab and Shift+Tab wrap between Cancel and Enable trusted app; Escape and Cancel close it and restore focus to the still-disabled Enable switch.+- `nav_label` is dropped by the Agent Server: `GET /api/canvas-extensions/installed` returns pages with only `id`, `title` and `path` (Agent Server 1.50.1), so the rail shows the page title ([OpenHands/software-agent-sdk#5501](https://github.com/OpenHands/software-agent-sdk/issues/5501)). Keep `F20.rail-label` failing until the backend returns it.- Install and update errors show the server's reason as sent, even when it says `Failed to fetch extension from <url>` (an unreachable Git source). A response with no reason in it, such as the ingress's `502` while the Agent Server is stopped, shows the generic `An error occurred`. Only a request that gets no response at all shows `Disconnected (check URL or network)…`; no `control-openhands` verb takes the ingress offline, so that state is not driven here.- The page states (empty, error, unsupported) and the `/extensions/...` unavailable card have no test ids; assert on text with `browser snapshot`.- The Agent Server fetches Git sources with a shallow clone and appends `.git` to the repo URL: a test Git server must speak smart HTTP and answer both `qa-repo` and `qa-repo.git`.@@ -96,3 +99,6 @@ Preconditions:- Stopping the Agent Server makes the next full page load show the **Manage backends** gate; always `restart` and `doctor` afterwards.- A trailing `browser errors --app-only` sweep shows a `400` per failed install by design; count only other errors.- `App backend view unavailable` on the demo page is the expected text while `GET /server_info` lists no `canvas_app_backend_bridge_v1` capability (Agent Server 1.53.0); it does not mean the app failed to load. Cloud backends and the no-backend state never get a backend view either.+- Known issue [OpenHands/software-agent-sdk#5648](https://github.com/OpenHands/software-agent-sdk/issues/5648) (`F20.app-backend-view`): app-backend WebSocket subprotocols are reported dropped. Recheck on a Linux Agent Server with configured public ingress and reviewed platform-matching backend artifacts, using a genuine admitted app-backend session and subprotocol negotiation. SDK 1.55.0 supports only Linux app-backend artifacts, so a native macOS server cannot drive this branch. A default demo without the bridge capability cannot exercise the proxied WebSocket path.+- Known issue [OpenHands/software-agent-sdk#5935](https://github.com/OpenHands/software-agent-sdk/issues/5935) (`F20.app-backend-view`): an explicit `OH_APP_BACKEND_PUBLIC_URL_IS_NONE=1` does not clear a bridge URL set in the Agent Server JSON configuration. Canvas can therefore be offered an app-backend view at the unwanted configured ingress. Recheck an owned bridge-capable server with a JSON public URL and that environment-null override, then inspect `/server_info` and the real app page; the default launcher has neither conflicting configuration source.+- Known issue [OpenHands/software-agent-sdk#5914](https://github.com/OpenHands/software-agent-sdk/issues/5914) (`F20.app-backend-view`): app-backend HTTP responses can carry both the backend's and the Agent Server's `Date` and `Server` headers, producing an invalid joined date and disclosing the backend stack. Recheck a reviewed HTTP app backend on a supported Linux Agent Server that sends these headers through a real admitted app-backend session. A default demo page without the bridge capability or a direct backend request cannot establish the proxied response.diff --git a/.agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md b/.agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.mdindex 6e9ddb0e9e..68e9d50b42 100644--- a/.agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md+++ b/.agents/skills/verify-openhands/references/feature-map/F21-automations-dashboard.md@@ -108,7 +108,7 @@ Preconditions:- **Open the detail page (`F21.open-detail`).** Kebab `role=list >> has-text=Export >> role=button[name="View"]`, `control-openhands browser click '[data-testid^="automation-list-row-"] >> has-text=QA_F21 Zulu >> role=link >> nth=0'`, `control-openhands browser click '[data-testid^="automation-card-"] >> has-text=QA_F21 Zulu >> role=heading'` (grid: switch first with `control-openhands browser click 'testid=automations-view-toggle'` and `control-openhands browser click 'testid=automations-view-toggle-grid'`; the later bullets use cards), and `control-openhands browser focus '[data-testid^="automation-card-"] >> has-text=QA_F21 Beta'` + `control-openhands browser press Enter` each make `control-openhands browser url` read `/automations/<that automation's id>`. Go back with `control-openhands browser back`.- **Git Sync button (`F21.git-sync-button`).** `control-openhands browser click 'testid=automations-git-sync'`, then `control-openhands browser url`: `/automations/git-sync` (F24).- **Needs attention, kebab Run now (`F21.overview-tiles`, `F21.run-now`, `F21.filters`).** Run `control-openhands browser goto /automations` (the Git Sync button left `/automations/git-sync`), import `qa-f21-fail.automation.json` as above, turn it on from its kebab, then `control-openhands browser click '[data-testid^="automation-card-"] >> has-text=QA_F21 Fail >> role=button[name="Automation actions"]'`, `control-openhands browser enabled 'role=list >> has-text=Export >> role=button[name="Run now"]'` (`true`), click it, `control-openhands browser wait-text 'Automation dispatched'`. The 1 s timeout run ends `FAILED` after about 3-4 minutes (check with `control-openhands api GET /api/automation/v1/<id>/runs`; the id is in `api GET /api/automation/v1`). Then `control-openhands browser reload`, `control-openhands browser text 'testid=overview-tile-needs-attention'` (`Needs attention 1 / Latest run failed`), and `control-openhands browser screenshot --feature F21.overview-tiles --name needs-attention`. The Fail card shows **Failed** with `Timed out: command timed out or was killed`, a red sparkline bar and `Recent success 0%`; Total runs is `2`. Open Filters and choose status `automations-filter-status-failing` (`testid=automations-filter-status >> testid=dropdown-trigger`, then `testid=automations-filter-status-failing`): only QA_F21 Fail is left. In the still-open popover click `testid=automations-filters-reset` (all three automations return), then close it with `control-openhands browser click 'role=heading[name="Dashboard"]'`: the next bullets need QA_F21 Fail and QA_F21 Beta visible.-- **Run now error (`F21.run-now-error`).** Arrange a stale card: with the dashboard open, run `control-openhands api DELETE /api/automation/v1/<QA_F21 Fail id> --write` and `control-openhands browser toasts --history --clear` (empties the toast history), then click `'[data-testid^="automation-card-"] >> has-text=QA_F21 Fail >> role=button[name="Run now"]'`, then `control-openhands browser toasts --history` and `control-openhands browser screenshot --feature F21.run-now-error --name error-toast`. The history holds exactly one toast, `Automation not found`, and no `Request failed with status code 404`. `browser errors --app-only` then lists the induced 404 on `/dispatch`. Expected: the card then drops out without a reload: `control-openhands browser wait '[data-testid^="automation-card-"] >> has-text=QA_F21 Fail' --state detached --timeout 10000` succeeds. Known failure: the wait times out and the card stays until `control-openhands browser reload` (#18062, see Gotchas).+- **Run now error (`F21.run-now-error`).** Arrange a stale card: with the dashboard open, run `control-openhands api DELETE /api/automation/v1/<QA_F21 Fail id> --write` and `control-openhands browser toasts --history --clear` (empties the toast history), then click `'[data-testid^="automation-card-"] >> has-text=QA_F21 Fail >> role=button[name="Run now"]'`, then `control-openhands browser toasts --history` and `control-openhands browser screenshot --feature F21.run-now-error --name error-toast`. The history holds exactly one toast, `Automation not found`, and no `Request failed with status code 404`. `browser errors --app-only` then lists the induced 404 on `/dispatch`. Expected: the card then drops out without a reload: `control-openhands browser wait '[data-testid^="automation-card-"] >> has-text=QA_F21 Fail' --state detached --timeout 10000` succeeds. Known failure: the wait times out and the card stays until `control-openhands browser reload` ([#18062](https://github.com/OpenHands/OpenHands/issues/18062), see Gotchas).- **Delete with confirmation (`F21.delete`).** Open Beta's kebab and click `role=list >> has-text=Export >> role=button[name="Delete"]`. `control-openhands browser snapshot 'role=dialog[name="Delete automation"]'` shows dialog `Delete automation` with `Close`, the heading, `Delete "QA_F21 Beta"? This cannot be undone.`, `Cancel` and `Delete`; focus starts on Cancel (`control-openhands browser eval "document.activeElement.textContent"` is `Cancel`) and `browser press Tab` / `Shift+Tab` cycle Cancel, Delete and Close inside the dialog. Take `control-openhands browser screenshot --feature F21.delete --name dialog`. `control-openhands browser click 'role=dialog[name="Delete automation"] >> role=button[name="Cancel"]'` closes it (`control-openhands browser count 'role=dialog[name="Delete automation"]'` is `0`) with the card kept. Reopen and close it with the X (`control-openhands browser click 'role=dialog[name="Delete automation"] >> role=button[name="Close"]'`), then reopen and `control-openhands browser press Escape`: the dialog count is `0` and the card count `1` each time. Reopen, then `control-openhands browser click 'role=dialog[name="Delete automation"] >> role=button[name="Delete"]'`, `control-openhands browser reload` and `control-openhands browser count '[data-testid^="automation-card-"] >> has-text=QA_F21 Beta'`. The count is `0`. Delete the remaining fixtures the same way (rows work too) until `testid=automations-empty` returns.- **Load more (`F21.load-more`).** Arrange 51 inert automations (event trigger that never fires): `for i in $(seq -w 1 51); do control-openhands api POST /api/automation/v1/preset/prompt --write --data "{\"name\":\"QA_F21 Page $i\",\"prompt\":\"QA paging fixture; never runs.\",\"trigger\":{\"type\":\"event\",\"source\":\"qa-f21\",\"on\":\"qa.f21.never\"}}"; done`. Then `control-openhands browser goto /automations`, `control-openhands browser wait 'role=button[name="Load more"]'`, row/card count `50`, `control-openhands browser click 'role=button[name="Load more"]'`, `control-openhands browser wait 'role=button[name="Load more"]' --state detached`, count `51`. The preset fixtures arrive enabled (`51 active`) but never fire. Keep them for Phone (it needs at least one automation) and clean up after it.- **Phone (`F21.phone`, `F21.subpage-nav`).** With at least one automation, run `control-openhands browser viewport phone`, `control-openhands browser bbox 'role=main'` (`insideViewport` `true`, `pageHorizontalOverflow` `false`), `control-openhands browser screenshot --feature F21.phone --name dashboard`, `control-openhands browser click 'testid=automations-navbar-mobile >> testid=automations-navigation-templates'` (URL `/automations/templates`) and back through `automations-navigation-list`. Open the first card's kebab and run `control-openhands browser bbox 'role=list >> has-text=Export'`: the menu is `insideViewport` `true` and sits above the kebab (flipped). Take `control-openhands browser screenshot --feature F21.phone --name kebab`, then `control-openhands browser press Escape` and `control-openhands browser viewport desktop`. Now clean up the paging fixtures with `control-openhands api DELETE /api/automation/v1/<id> --write` for each `QA_F21 Page` id from `control-openhands api GET '/api/automation/v1?limit=100'` (UI deletion is covered above) until `total` is `0`.@@ -122,7 +122,7 @@ Preconditions:- The setup page's success toast (`Automation updated`) covers its toolbar's Test, Save and Close and does not time out while the pointer rests on it; a click there fails with "subtree intercepts pointer events". Move the pointer away (`browser hover` on a form field), wait about 5 s, then click.- `testid=dropdown-trigger` is reused by the backend selector, the Filters button and every inner dropdown. Always scope it (`testid=automations-filters >> testid=dropdown-trigger >> nth=0` for the Filters button, `testid=automations-filter-status >> testid=dropdown-trigger` for a field).- The Filters popover stays open while choices change and closes only on a click outside it or on the Filters button; Escape does not close it. Open, it covers the right part of the list (in list view the first rows' Play buttons): close it with a click on `role=heading[name="Dashboard"]` before clicking row or card controls.-- A card whose automation was deleted elsewhere stays on the dashboard after its Run now fails with `Automation not found`. Expected: the card drops out once the dispatch returns 404. Known failure: nothing refetches the list after the failed dispatch, so the card stays until a reload or another list refetch (still there 2 minutes later) (#18062).+- A card whose automation was deleted elsewhere stays on the dashboard after its Run now fails with `Automation not found`. Expected: the card drops out once the dispatch returns 404. Known failure: nothing refetches the list after the failed dispatch, so the card stays until a reload or another list refetch (still there 2 minutes later) ([#18062](https://github.com/OpenHands/OpenHands/issues/18062)).- The delete confirmation has no test ids; scope it as the named dialog `role=dialog[name="Delete automation"]`. It opens with focus on Cancel.- Delete closes the dialog at once and shows no success toast; confirm with a reload. Turn off on the dashboard never asks for confirmation (home cards do).- The kebab menu has no test ids; `role=list >> has-text=Export` is the reliable scope. A card's own **Run now** has the same accessible name as the kebab item, so scope Play to the card.@@ -137,3 +137,5 @@ Preconditions:- `service stop automation` stops only this run's automation service; `restart` brings the whole stack back with the same state and the browser on the same page. Run `control-openhands doctor` afterwards.- The **Automation dispatched** toast lasts only a few seconds: run `browser wait-text 'Automation dispatched'` right after the click, not after a long `--observe` window.- `browser screenshot --full-page` at phone width does not capture below the fold, because the page scrolls inside the main pane; screenshot sections or use `bbox`.+- Known issue [OpenHands/automation#246](https://github.com/OpenHands/automation/issues/246) (`F21.backend-unavailable`, `F21.list-error`): The report says only that local automation returns 404. It omits the URL, UI action, release and launcher/service setup, so a normal successful dashboard visit would not reproduce or resolve its unknown branch. Ask for those exact prerequisites before writing a failing recipe.+- Known issue [OpenHands/automation#347](https://github.com/OpenHands/automation/issues/347) (`F21.card-grid`, `F21.overview-tiles`): per-automation run-summary reads lack an aggregate contract, and the report describes a 30 s shared-resource stall with HTTP 500 plus SQLite pool settings that do not apply. A failing recipe needs the original database/lock workload and redacted service-log/HAR timing evidence, which the report withholds. Successful summary reads under the Canvas request throttle do not establish those backend contracts.diff --git a/.agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md b/.agents/skills/verify-openhands/references/feature-map/F22-automation-creation.mdindex 008d261b9d..fe28a19ba1 100644--- a/.agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md+++ b/.agents/skills/verify-openhands/references/feature-map/F22-automation-creation.md@@ -109,6 +109,8 @@ Preconditions:- **Phone (`F22.phone`).** Run `control-openhands browser viewport phone`, `control-openhands browser goto /automations/templates`, `control-openhands browser bbox 'testid=recommended-automations-section'` (`pageHorizontalOverflow` `false`) and `browser screenshot --feature F22.phone --name templates`. Click `testid=recommended-automation-card-custom-automation`, then `testid=setup-continue-button`; `browser bbox 'testid=setup-dialog'` is `insideViewport` `true` with no page overflow (`--name setup`). Close it, `control-openhands browser goto /automations`, open Add Automation → Import automation, upload `qa-f22-import.automation.json`; `browser bbox 'testid=import-automation-modal'` is `insideViewport` `true` (`--name import-preview`). Close with `testid=import-automation-modal-close`, then `control-openhands browser viewport desktop`.- **Cleanup.** Deleting an automation is F21's flow; here remove the fixtures with `control-openhands api DELETE /api/automation/v1/<id> --write` for each `QA_F22 …` (Custom, Plugin, Tarball, Imported, View, Agent) and `Daily news digest` id from `control-openhands api GET '/api/automation/v1?limit=100'`. Delete `OPENHANDS_URL` in Settings → Secrets (`control-openhands browser click 'testid=secret-item >> has-text=OPENHANDS_URL >> testid=delete-secret-button'`, then `testid=confirmation-modal >> testid=confirm-button`, as in F14.delete) unless the next recipe wants it. The fallback and setup conversations stay in the conversation list (F04 deletes them). With zero automations, `control-openhands browser goto /automations` and `control-openhands browser click 'testid=automations-empty >> testid=automations-create-automation' --expect-url '/automations/setup'` opens the agent-assisted setup page.+- **Malformed catalog entry (`F22.template-card`, `F22.templates-page`).** Archive the frozen target into an isolated private checkout, link unchanged dependencies except a private copy of the pinned `@openhands/extensions`, and remove only `requires` from `automations/catalog/custom-automation/manifest.json`. Leave Canvas source and its lockfile unchanged; run `npm run build:app` only in that copy. Serve its flat `build/` directory against an owned healthy backend, then open `/automations/templates` through `browser goto <fixture-origin>/automations/templates --allow-external`. Expected: the malformed entry degrades without crashing the valid catalog. Known failure, reproduced 2026-10-12 on `fa1f724`: the whole page renders **Application Error**, `TypeError: Cannot read properties of undefined (reading 'integrations')` ([#16277](https://github.com/OpenHands/OpenHands/issues/16277)). Record this as catalog-data fault injection, then stop/remove only the private fixture frontend/build; never modify the shared package or build.+## Gotchas- Expected: a completed draft test updates the open Test runs row to Successful automatically. Known failure (reproduced again 2026-10-13 on a desktop prompt draft: run COMPLETED at 08:13:39, row still Pending after a further 45 s wait; verified 2026-10-10, fresh Local stacks, Agent Server 1.54.0/automation 1.19.3): callback-only script tests and genuine DeepSeek prompt/plugin tests report COMPLETED while the open row stays Pending. The desktop prompt stayed Pending through a 45-second wait; the phone plugin capture was more than 201 seconds after completion. Reload and View test runs recover Successful ([#18263](https://github.com/OpenHands/OpenHands/issues/18263)). Ordinary final-automation Run now polling passes; reload recovery does not waive the draft-Test failure.@@ -117,7 +119,7 @@ Preconditions:- In the desktop split view (form beside the conversation) the Frequency strip scrolls horizontally inside the narrower form and the selected **Custom** button sits off-screen; scroll it into view before reading it.- Dashboard draft Delete asks in the named dialog `role=dialog[name="Delete draft?"]` (F21.saved-drafts). Deleting a draft also removes its materialized test automation.- A card click and a deep link are different paths: from the grid, GitHub/Slack-only responders (`qa-changes`, `github-agents-md-maintainer`, `slack-channel-monitor`, …) first show the responder choice and then the GitHub/Slack install dialog, so their setup form is reachable locally only by URL (`/automations/new/<id>`).-- A card's `N MCPs to connect before launch` pill counts optional integrations too: Custom automation shows `1 MCPs to connect before launch` for GitHub, yet opens its setup at once because GitHub is optional there (#17959).+- A card's `N MCPs to connect before launch` pill counts optional integrations too: Custom automation shows `1 MCPs to connect before launch` for GitHub, yet opens its setup at once because GitHub is optional there ([#17959](https://github.com/OpenHands/OpenHands/issues/17959)).- The featured-section copy says picking a card opens "a pre-filled conversation"; most featured cards open a setup form instead.- The setup dialog's comboboxes (Action, Agent profile, Timezone, LLM profile) are autocomplete inputs: click the field and pick the option in the same breath, scoped as `role=listbox >> role=option[name="…"]`. A `browser snapshot` between the two closes the listbox (reproduced twice), and the option click then times out after 30 s; click the field again.- Service preflight runs 400 ms after a blur, not on every keystroke; `browser press Tab` then `browser wait` for the `-error` element.@@ -129,4 +131,8 @@ Preconditions:- **Continue with local setup** silently creates the `OPENHANDS_URL` secret (value: the app origin) the first time; F14's list then has an extra row.- The **Install** button in a template's install dialog verifies the MCP connection before saving, so dummy credentials cannot complete the queue; nothing is saved on failure.- Creating the first automation ticks **Schedule a task** in the sidebar Getting-started checklist (F02); a run that checks the checklist should do so before this file.-- Known issue #17959: a template search with no matches leaves a blank area with no message (`F22.templates-search`).+- Known issue [#17959](https://github.com/OpenHands/OpenHands/issues/17959): a template search with no matches leaves a blank area with no message (`F22.templates-search`).+- Known issue [#17436](https://github.com/OpenHands/OpenHands/issues/17436) (`F22.setup-review-confirm`): Prompt-preset automations created after the Canvas 1.17.0 upgrade fail on the first conversation events request while older tarballs work, so the UI shows Failed runs. Recheck with a real LLM key and the reported automation 1.11.1/Agent Server 1.46.0 deployment plus pre-upgrade and finish-tool-wired tarballs; retain this version-specific report separately from the plugin-preset Cloud failure in #16611.+- Known issue [#15509](https://github.com/OpenHands/OpenHands/issues/15509) (`F22.setup-tarball-create`): A native Windows event-triggered automation fails while uploading its workspace tarball with `Path must be absolute`. Recheck a Windows local stack and real event delivery, then observe its failed run/upload; a POSIX tarball upload is not a Windows path-validation check. This run has neither Windows nor the webhook fixture.+- Known issue [#16611](https://github.com/OpenHands/OpenHands/issues/16611) (`F22.setup-plugin-create`): a minimal plugin-preset Cloud automation fails on its in-sandbox conversation/events request with HTTP 500, while the identical prompt preset works. Live prerequisites: an authenticated Cloud account, a real model key, and a reviewed no-op plugin fixture for the paired cases. The reported cost-zero error is visible in run logs, but a local prompt-only check cannot establish the Cloud plugin-loading behavior.+- Known issue [OpenHands/software-agent-sdk#5729](https://github.com/OpenHands/software-agent-sdk/issues/5729) (`F22.setup-repository-list`): prompt and plugin preset workers call `RemoteWorkspace.clone_repos`, which asks for lowercase provider-token secret names. A private repository can therefore fail to clone when only the documented uppercase `GITHUB_TOKEN` (or GitLab/Bitbucket equivalent) exists. Recheck an owned preset with a private repository and only its uppercase secret, using an authorized disposable repository or a reviewed token-gated local Git fixture; a public repository and the Canvas Agent Server's separate Git-provider service do not exercise that lookup.diff --git a/.agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md b/.agents/skills/verify-openhands/references/feature-map/F23-automation-detail.mdindex 5d272fff44..e4392a0ea1 100644--- a/.agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md+++ b/.agents/skills/verify-openhands/references/feature-map/F23-automation-detail.md@@ -54,7 +54,7 @@ Source: `src/routes/automation-detail.tsx`, `src/components/features/automations- `F23.edit-test`: on the edit page **Test** dispatches the automation as last saved (`POST /api/automation/v1/<id>/dispatch`, no PATCH first) and shows `Test run started.` on the form; unsaved edits are not part of that run. Whether it should test the current form is a product decision (Gotchas).- `F23.edit-dirty-state`: the edit page shows no save state for unsaved edits (no **Unsaved changes**, which create mode shows, F28); the label appears only while saving (**Saving…**) and after (**Saved just now** or **Save failed**). **Close** discards the edits.- `F23.edit-schedule`: Frequency offers Once, Hourly, Daily, Weekdays (Mon–Fri), Weekly and Custom; Weekly adds **Day of week** and **At**; they are saved as a cron expression and read back after a reload.-- `F23.edit-time-cleared`: clearing **At** on a preset schedule and saving should be refused with a message instead of being dropped or replaced. Known failure (reproduced 2026-10-13, #17948 item, new symptom since #17688): Save silently stores midnight (`0 0 * * 3`) and toasts **Automation updated**; Save draft in the create flow stores `0 0 * * *` the same way. Open PR #18260 (head `0f812ac68902d2c71896568311bffb8aa6f216a6`, checked 2026-10-15) targets the live setup panel: it rejects an invalid or cleared time before Save and Save draft, with an edit-mode regression test. It is not merged; the known failure above remains the observed behavior on this map’s baseline.+- `F23.edit-time-cleared`: clearing **At** on a preset schedule and saving should be refused with a message instead of being dropped or replaced. Known failure (reproduced 2026-10-13, [#17948](https://github.com/OpenHands/OpenHands/issues/17948) item, new symptom since #17688): Save silently stores midnight (`0 0 * * 3`) and toasts **Automation updated**; Save draft in the create flow stores `0 0 * * *` the same way. Open PR #18260 (head `0f812ac68902d2c71896568311bffb8aa6f216a6`, checked 2026-10-15) targets the live setup panel: it rejects an invalid or cleared time before Save and Save draft, with an edit-mode regression test. It is not merged; the known failure above remains the observed behavior on this map’s baseline.- `F23.edit-cron`: a custom cron schedule shows Frequency **Custom** with an editable **Custom** cron field and a timezone select.- `F23.edit-event-readonly`: for an event automation the edit page shows Event selected with Event source, event key and filter controls (editable now, no longer read-only text) and no schedule fields.- `F23.edit-save-error`: a failed save shows exactly one error toast with the API message (`Automation not found` for an automation deleted meanwhile), the save state reads **Save failed** and the edits stay on the page; the toolbar **Test** on that page also toasts `Automation not found`.@@ -98,20 +98,25 @@ Preconditions:- **Validation (`F23.edit-validation`).** On the edit page run `control-openhands browser fill 'testid=automation-setup-prompt' ''`, `control-openhands browser network --clear` and `control-openhands browser click 'testid=automation-setup-save'`: `control-openhands browser toasts --history` includes the toast `Add a prompt before continuing.` and `browser network --filter 'automation/v1'` lists no request. Refill the prompt. On a Custom-cron automation (QA_F23 Fail) fill `testid=automation-setup-custom-schedule` with `not a cron` (then `0 0 31 2 *`, then a valid cron with `testid=automation-setup-timeout` `1801`), click Save and read `control-openhands browser toasts --history` after `browser toasts --history --clear`. Each sends a `PATCH` that returns `422` (`browser network --filter 'automation/v1'`), and the history holds exactly one toast with the API message: `Value error, Invalid cron expression: not a cron`, `Value error, Cron expression cannot produce any future fire times: 0 0 31 2 *` and `Value error, timeout must not exceed 1800 seconds (30 minutes)`; `api GET /api/automation/v1/<Fail id>` keeps `0 0 1 1 *` and timeout `600`. Before each new click move the pointer off the toast (`browser hover 'testid=automation-setup-name'`, about 5 s; see Gotchas).- **No-change Save (`F23.edit-save`).** Open Edit, `control-openhands browser network --clear`, click `testid=automation-setup-save`, then `control-openhands browser network --filter 'automation/v1'`: one `PATCH` with the unchanged values and the toast `Automation updated` (the old modal sent nothing).- **Weekly schedule, model and timeout (`F23.edit-schedule`, `F23.edit-save`).** On the edit page run `control-openhands browser click 'testid=automation-setup-frequency-weekly'`, `control-openhands browser select 'testid=automation-setup-weekday' 3` (Wednesday; options `0:Sunday` … `6:Saturday`), `control-openhands browser fill 'testid=automation-setup-time' 10:30`, `control-openhands browser click 'testid=automation-setup-model'`, `control-openhands browser click 'testid=automation-setup-model-option-deepseek-pro'`, `control-openhands browser fill 'testid=automation-setup-timeout' 300` and `control-openhands browser screenshot --feature F23.edit-schedule --name weekly-form`. Then `control-openhands browser network --clear`, `control-openhands browser click 'testid=automation-setup-save' --observe 'testid=automation-setup-save-state' --observe-ms 2000` (`Saving…` → `Saved just now`) and `control-openhands browser toasts --history` (`Automation updated`). `browser network --bodies --filter 'automation/v1'` shows one `PATCH /api/automation/v1/<id>` with `schedule` `30 10 * * 3`, `model` `deepseek-pro` and `timeout` `300`. After `control-openhands browser reload` the weekday is `3`, At `10:30`, the model `deepseek-pro` and Timeout `300`; the detail page Configuration reads `Schedule` `30 10 * * 3` and `LLM profile` `deepseek-pro`.-- **Cleared time (`F23.edit-time-cleared`).** On the weekly edit page `control-openhands browser fill 'testid=automation-setup-time' ''`, `control-openhands browser network --clear`, click Save. Expected: an inline message and nothing saved. Known failure (reproduced 2026-10-13, #17948 item, new symptom since #17688): `browser network --bodies --filter 'automation/v1'` shows a `PATCH` with `schedule` `0 0 * * 3`, the toast reads `Automation updated` and the automation now runs at midnight. Save draft in the create flow (`/automations/setup`) stores `0 0 * * *` the same way; open PR #18260 targets both paths in `automation-setup-panel.tsx` at head `0f812ac68902d2c71896568311bffb8aa6f216a6` (checked 2026-10-15), but is not merged.+- **Cleared time (`F23.edit-time-cleared`).** On the weekly edit page `control-openhands browser fill 'testid=automation-setup-time' ''`, `control-openhands browser network --clear`, click Save. Expected: an inline message and nothing saved. Known failure (reproduced 2026-10-13, [#17948](https://github.com/OpenHands/OpenHands/issues/17948) item, new symptom since #17688): `browser network --bodies --filter 'automation/v1'` shows a `PATCH` with `schedule` `0 0 * * 3`, the toast reads `Automation updated` and the automation now runs at midnight. Save draft in the create flow (`/automations/setup`) stores `0 0 * * *` the same way; open PR #18260 targets both paths in `automation-setup-panel.tsx` at head `0f812ac68902d2c71896568311bffb8aa6f216a6` (checked 2026-10-15), but is not merged.- **Custom cron (`F23.edit-cron`).** Open QA_F23 Fail and Edit: Frequency **Custom** is selected and `control-openhands browser value 'testid=automation-setup-custom-schedule'` is `0 0 1 1 *` beside the timezone select (`UTC`). Fill `0 0 2 1 *` and `testid=automation-setup-timeout` with `1` (this makes the next run fail), Save and `browser toasts --history` (`Automation updated`); after a reload of the detail page the Configuration shows `0 0 2 1 *`. Invalid crons: see Validation.- **Failed run, its row and Debug with OpenHands (`F23.run-status-polling`, `F23.run-no-conversation`, `F23.run-logs-modal`, `F23.debug-with-openhands`).** On QA_F23 Fail (timeout `1` from the previous bullet) click `role=button[name="Run now"]` and `control-openhands browser wait-text 'Automation dispatched'`. The run stays Running for about 3-4 minutes and ends `FAILED` (poll `control-openhands api GET /api/automation/v1/<id>/runs`; drive the event automation bullets meanwhile). Then `control-openhands browser goto /automations`, open QA_F23 Fail, `control-openhands browser wait 'testid=automation-activity-log >> testid=run-status-icon-failed'` and `control-openhands browser text 'testid=automation-activity-log'`: the newest row reads `<time>`, `Timed out: command timed out or was killed`, `Failed`. Expected beside the time: `No Conversation` (this is a prompt automation); today it reads `No conversation — this run executed a script. Use View logs for its output.` Run `control-openhands browser click 'testid=automation-activity-log >> title=View logs >> nth=0'` and `control-openhands browser snapshot 'role=dialog'`: `Run` `FAILED`, `Task` `No task outcome reported.`, `System` `Error: Timed out: command timed out or was killed` and `Status detail: ...`, the Output tab with the `[setup] Fetching SDK version ...` lines, and button `Debug with OpenHands`. Screenshot with `--feature F23.debug-with-openhands --name failed-logs`. `control-openhands browser click 'testid=debug-automation-button' --expect-url '/conversations/'` opens a new conversation; `control-openhands conversation events <conversation id from browser url> --kinds MessageEvent` starts `The scheduled automation "QA_F23 Fail" failed during a run. Please investigate the error and fix the root cause.`, and `control-openhands browser eval "document.body.innerText.includes('Timed out: command timed out or was killed')"` is `true` (the page also shows the run id). The agent then works on that prompt with the active profile.- **Task outcome badge (`F23.run-task-outcome`).** Arrange (API, arrange only) `control-openhands api POST /api/automation/v1/preset/prompt --write --data '{"name":"QA_F23 Blocked","prompt":"Do not run any tools. Call the finish tool right away with status \"blocked\" and outcome_summary \"QA blocked on purpose\".","trigger":{"type":"cron","schedule":"0 0 1 1 *","timezone":"UTC"},"enabled":true}'` (a good thing to do while the Fail run is still running). Open QA_F23 Blocked, click `role=button[name="Run now"]`, and poll `control-openhands browser text 'testid=automation-activity-log'` until it leaves Running (about 35 s): `<time>`, `QA blocked on purpose`, a cost such as `$0.0016`, `Blocked`; `control-openhands browser testids 'testid=automation-activity-log'` lists `run-status-icon-warning` (not `-completed`) although `api GET /api/automation/v1/<id>/runs` has `status` `COMPLETED` and `run_metadata.finish_tool_response.status` `blocked`. `control-openhands browser click 'testid=automation-activity-log >> title=View logs'` and `browser snapshot 'role=dialog'`: `Run` `COMPLETED`, `Task` `Blocked` / `QA blocked on purpose` / `Task metadata` `{ "message": ... }`. Screenshot with `--feature F23.run-task-outcome --name blocked`, Escape. `partial_success` (**Partial**, `run-status-icon-warning`) and `unknown` (**Needs review**, `run-status-icon-needs-review`) follow the same path with the same prompt shape. The normal automation 1.19.0 prompt/plugin presets configure SDK 1.53.0 FinishTool with the required `TaskOutcome.status`, limited to `success`, `partial_success`, `blocked`, `failed` and `unknown`; SDK custom response schemas are a separate producer boundary. Its normal prompt path therefore does not establish the new fallback for a finish response without a string status. That positive boundary needs a genuine completed run from a legacy or custom-schema producer whose real finish arguments have that shape; a run with no finish metadata checks only the ordinary Completed fallback. Arbitrary unrecognised strings and summary retention on a failed run need the corresponding genuine producer too. Until available, record those checks as blocked and do not treat injected run metadata or mocked LLM output as live proof (#17682).- **Event automation (`F23.prompt-collapse`, `F23.configuration-event`, `F23.plugins-repo`, `F23.export`).** Open QA_F23 Event. `control-openhands browser bbox 'testid=automation-prompt-content'` is 240 px high, `testid=automation-prompt-fade` is present and `control-openhands browser text 'testid=automation-prompt-toggle'` is `View More`; screenshot with `--feature F23.prompt-collapse --name collapsed`. Click the toggle: height 360, no fade, text `View Less`; click again to collapse. `control-openhands browser text 'role=heading[name="Configuration"] >> xpath=ancestor::div[2]'` reads `Repositories qa-example/qa-repo main`, `Trigger Event`, `Event Source github`, `Event Type pull_request.opened`, `Filter pull_request.draft == `false` && pull_request.base.ref == re…`, `Show more`, `LLM profile Active profile`. `control-openhands browser click 'role=button[name="Show more"]'` shows the full filter and `control-openhands browser attr 'role=button[name="Show less"]' aria-expanded` is `true`. `control-openhands browser text 'testid=automation-repository'` is `qa-example/qa-repo` / `main` (the fixture's `https://github.com/qa-example/qa-repo` as `owner/repo`, with its branch badge) and `control-openhands browser text 'role=heading[name="Plugins"] >> xpath=ancestor::div[2]'` is `Plugins` / `github:qa-example/qa-plugin`; screenshot with `--feature F23.plugins-repo --name event-repo-plugins`. Export keeps them: `control-openhands browser click 'role=button[name="Automation actions"]'`, `control-openhands browser click 'role=list >> has-text=Tarball >> role=button[name="Export"]'` and `control-openhands browser downloads --last 1 --contains '"branch": "main"'` (`qa-f23-event.automation.json`, `contains` `true`); `--contains '"repository": "https://github.com/qa-example/qa-repo"'` and `--contains 'github:qa-example/qa-plugin'` are `true` too (the `--inspect` head ends inside the long prompt).- **Event trigger in Edit (`F23.edit-event-readonly`).** On QA_F23 Event open Edit: Event is selected, `browser testids 'testid=automation-setup-form' --filter event` lists `automation-setup-event-source` (GitHub), `automation-setup-event-source-toggle`, `automation-setup-event-public-url-notice`, `automation-setup-event-key`, `automation-setup-event-filter` and `automation-setup-event-filter-remove` (the fixture has a filter), and there are no Frequency controls for a schedule. The trigger is editable here (it was read-only text in the old modal). Screenshot with `--feature F23.edit-event-readonly --name event`, then Close.- **Phone (`F23.phone`).** Run `control-openhands browser viewport phone`, open QA_F23 Pong, `control-openhands browser bbox 'role=heading[name="QA_F23 Pong"] >> xpath=ancestor::div[3]'` (`insideViewport` `true`, `pageHorizontalOverflow` `false`) and `control-openhands browser screenshot --feature F23.phone --name detail` (the name and Run now wrap). Open Edit (kebab, `--expect-url '/automations/setup'`), then `control-openhands browser bbox 'testid=automation-setup-save'` and `... 'testid=automation-setup-close'` (both `insideViewport` `true`, Save at `y` about 65) and `control-openhands browser bbox 'testid=automation-setup-form'` (`pageHorizontalOverflow` `false`); screenshot with `--feature F23.phone --name edit-setup-page`. Close and `control-openhands browser viewport desktop`.-- **Run now and save errors (`F23.run-now-error`, `F23.edit-save-error`).** Arrange a stale page: open QA_F23 Fail, then `control-openhands api DELETE /api/automation/v1/<Fail id> --write`. Run `control-openhands browser toasts --history --clear` (empties the toast history), click `role=button[name="Run now"]`, then `control-openhands browser toasts --history` and `control-openhands browser screenshot --feature F23.run-now-error --name toasts`: exactly one toast, `Automation not found`, and no `Request failed with status code 404`. Expected: the page then shows **Automation not found**: `control-openhands browser wait 'role=heading[name="QA_F23 Fail"]' --state detached --timeout 15000` succeeds (the not-found state takes about 7 s, as in States) and `control-openhands browser snapshot 'role=paragraph >> text=Automation not found >> xpath=..'` lists link `Back to Automations` (do not use `wait-text 'Automation not found'` here: the toast has the same text). Once the heading is gone, run `control-openhands browser toasts --history` again: it still holds only `Automation not found` (a late second toast, `Request failed with status code 404`, would come from the page's own refetch of the deleted automation). Known failure: the wait times out and the page keeps heading `QA_F23 Fail`, `Active` and an enabled **Run now** (`control-openhands browser enabled 'role=button[name="Run now"]'` is `true`) until a reload (#18062). The save error does not reuse that page (it is gone once #18062 is fixed): arrange a second fixture, `control-openhands api POST /api/automation/v1/preset/prompt --write --data '{"name":"QA_F23 Stale","prompt":"Reply with the single word: pong.","trigger":{"type":"cron","schedule":"0 0 1 1 *","timezone":"UTC"},"enabled":false}'` (`body.id` is `<stale-id>`), open QA_F23 Stale, its kebab and `role=list >> has-text=Tarball >> role=button[name="Edit"]` (`--expect-url '/automations/setup'`). With the edit page open, `control-openhands api DELETE /api/automation/v1/<stale-id> --write` and `control-openhands browser toasts --history --clear`. Fill `testid=automation-setup-name` with `QA_F23 Stale renamed` and click `testid=automation-setup-save`: the page keeps the edit (`browser value 'testid=automation-setup-name'` still `QA_F23 Stale renamed`) and `browser toasts --history` holds exactly one toast, `Automation not found`. Screenshot with `--feature F23.edit-save-error --name toast`. Move the pointer off the toast (`browser hover 'testid=automation-setup-name'`, about 5 s), `browser toasts --history --clear` and click `testid=automation-setup-test`: again exactly one toast, `Automation not found`. Close. `browser errors --app-only` lists the induced `POST /api/automation/v1/<Fail id>/dispatch`, `PATCH /api/automation/v1/<stale-id>` and `POST /api/automation/v1/<stale-id>/dispatch` 404s; both fixtures are already deleted.+- **Run now and save errors (`F23.run-now-error`, `F23.edit-save-error`).** Arrange a stale page: open QA_F23 Fail, then `control-openhands api DELETE /api/automation/v1/<Fail id> --write`. Run `control-openhands browser toasts --history --clear` (empties the toast history), click `role=button[name="Run now"]`, then `control-openhands browser toasts --history` and `control-openhands browser screenshot --feature F23.run-now-error --name toasts`: exactly one toast, `Automation not found`, and no `Request failed with status code 404`. Expected: the page then shows **Automation not found**: `control-openhands browser wait 'role=heading[name="QA_F23 Fail"]' --state detached --timeout 15000` succeeds (the not-found state takes about 7 s, as in States) and `control-openhands browser snapshot 'role=paragraph >> text=Automation not found >> xpath=..'` lists link `Back to Automations` (do not use `wait-text 'Automation not found'` here: the toast has the same text). Once the heading is gone, run `control-openhands browser toasts --history` again: it still holds only `Automation not found` (a late second toast, `Request failed with status code 404`, would come from the page's own refetch of the deleted automation). Known failure: the wait times out and the page keeps heading `QA_F23 Fail`, `Active` and an enabled **Run now** (`control-openhands browser enabled 'role=button[name="Run now"]'` is `true`) until a reload ([#18062](https://github.com/OpenHands/OpenHands/issues/18062)). The save error does not reuse that page (it is gone once [#18062](https://github.com/OpenHands/OpenHands/issues/18062) is fixed): arrange a second fixture, `control-openhands api POST /api/automation/v1/preset/prompt --write --data '{"name":"QA_F23 Stale","prompt":"Reply with the single word: pong.","trigger":{"type":"cron","schedule":"0 0 1 1 *","timezone":"UTC"},"enabled":false}'` (`body.id` is `<stale-id>`), open QA_F23 Stale, its kebab and `role=list >> has-text=Tarball >> role=button[name="Edit"]` (`--expect-url '/automations/setup'`). With the edit page open, `control-openhands api DELETE /api/automation/v1/<stale-id> --write` and `control-openhands browser toasts --history --clear`. Fill `testid=automation-setup-name` with `QA_F23 Stale renamed` and click `testid=automation-setup-save`: the page keeps the edit (`browser value 'testid=automation-setup-name'` still `QA_F23 Stale renamed`) and `browser toasts --history` holds exactly one toast, `Automation not found`. Screenshot with `--feature F23.edit-save-error --name toast`. Move the pointer off the toast (`browser hover 'testid=automation-setup-name'`, about 5 s), `browser toasts --history --clear` and click `testid=automation-setup-test`: again exactly one toast, `Automation not found`. Close. `browser errors --app-only` lists the induced `POST /api/automation/v1/<Fail id>/dispatch`, `PATCH /api/automation/v1/<stale-id>` and `POST /api/automation/v1/<stale-id>/dispatch` 404s; both fixtures are already deleted.- **Load more runs (`F23.activity-log-more`).** Not run: it needs more than 20 runs of one automation (about 35 s and one model call each). With them, `control-openhands browser click 'testid=automation-activity-log >> role=button[name="Load more"]'` grows the list from 20 to 40 rows.- **Script automation (`F23.script-section`).** Arrange a script automation through F22's bundle entry (UI arrange; F22 owns the creation proof): `control-openhands browser goto /automations/templates`, `control-openhands browser click 'testid=recommended-automation-card-news-digest' --expect-url '/automations/new/news-digest'`, `control-openhands browser fill 'testid=setup-field-schedule' '0 0 1 1 *'`, `control-openhands browser click 'testid=setup-continue-button'`, `control-openhands browser wait 'testid=setup-review' --timeout 10000`, `control-openhands browser click 'testid=setup-continue-button' --expect-url '/automations/[0-9a-f-]{8,}' --timeout 60000`. The detail page of **Daily news digest** shows `role=heading[name="Script"]` and no Prompt (`browser count 'role=heading[name="Prompt"]'` is `0`); `control-openhands browser text 'testid=automation-script-entrypoint'` is `python3 main.py`, and `browser eval "Array.from(document.querySelectorAll('[data-testid=automation-script-file]')).map(e=>e.innerText.split('\n').slice(0,2).join(' | '))"` lists `main.py | Entrypoint` first, then `config.json` (both with line-numbered source). Screenshot with `--feature F23.script-section --name news-digest`. `automation-script-unavailable` (bundle unreadable) was not reached. Do not Run now: the script fetches public feeds and starts a model-backed conversation. Delete it in the Delete bullet.- **Edit-mode Test and unsaved edits (`F23.edit-test`, `F23.edit-dirty-state`).** Open QA_F23 Event and its kebab **Edit** (`--expect-url 'automationId='`). Fill `testid=automation-setup-name` with `QA_F23 Unsaved`: `control-openhands browser count 'testid=automation-setup-save-state'` is `0`. `control-openhands browser network --clear`, click `testid=automation-setup-test`; `control-openhands browser network --filter 'automation/v1'` shows `POST …/<id>/dispatch 201` and no `PATCH`, and the form shows `Test run started.` Move the pointer off the toast (`control-openhands browser hover 'testid=automation-setup-prompt'`) and click `testid=automation-setup-close` (`--expect-url '/automations$'`): the card still reads `QA_F23 Event`. The dispatched run is not checked here; Delete below removes it with the automation.- **Delete (`F23.delete`).** Open QA_F23 Event, click the kebab and `role=list >> has-text=Tarball >> role=button[name="Delete"]`. `control-openhands browser snapshot 'role=dialog[name="Delete automation"]'` shows `Delete "QA_F23 Event"? This cannot be undone.`, `Cancel` and `Delete` (the same dialog as F21.delete). `control-openhands browser click 'role=dialog[name="Delete automation"] >> role=button[name="Cancel"]'` keeps the page (`control-openhands browser count 'role=dialog[name="Delete automation"]'` is `0`). Reopen and `control-openhands browser click 'role=dialog[name="Delete automation"] >> role=button[name="Delete"]' --expect-url '/automations(\?|$)'`; `control-openhands api GET /api/automation/v1` no longer lists it. Delete QA_F23 Pong, QA_F23 Blocked and Daily news digest the same way (and QA_F23 Fail if the error bullet was skipped); `api GET /api/automation/v1` shows `"total": 0`.- **Errors.** After each group `control-openhands browser errors --app-only` shows no page errors; the only app errors are the induced ones (404s for the unknown id and the deleted fixture, 502s while the service is stopped).+- **UTF-8 script member (`F23.script-section`).** Arrange an inactive annual-cron script automation through the real upload/create APIs with a two-file UTF-8 USTAR bundle (`tarfile.USTAR_FORMAT`, names `café.py` and `main.py`) and entrypoint `python3 café.py`; do not use a PAX-only member name. Open `/automations/<id>` and inspect `automation-script-entrypoint`, both `automation-script-file` headings, and the screenshot. Expected: `café.py` keeps its name, receives the **Entrypoint** badge and sorts first. Known failure, reproduced 2026-10-12 on `fa1f724`: the command is correct, but the file heading is `café.py` without the badge; the ASCII control reads normally ([#17704](https://github.com/OpenHands/OpenHands/issues/17704)). No run or model is needed.+- **Live free-form phase (`F23.run-status-polling`, `F23.activity-log`).** Upload an owned bounded script, create it active with the inert yearly cron, open its detail page, and click **Run now**. While genuinely running, compare `api GET /api/automation/v1/<id>/runs` with `browser snapshot 'testid=automation-activity-log'`. Expected: a run's `current_phase` renders as its live/last phase when structured fields are absent. Known failure, reproduced 2026-10-12 against automation 1.19.3: the API says `Starting automation` with no `phase_code`/`phase_label`, but Activity Log shows only **Running** ([#17121](https://github.com/OpenHands/OpenHands/issues/17121)). Keep the worker bounded and clean up only its owned command; the script needs no model.+- **Custom nested setup (`F23.script-section`, `F23.run-logs-modal`; creation in F22).** Arrange a reviewed tarball containing `scripts/setup.sh` that writes `.qa-setup-marker` and `main.py` that asserts that marker exists. Upload with `api POST /api/automation/v1/uploads?name=qa-nested-setup --write --data-file <tarball> --content-type application/gzip`; create with `entrypoint: python3 main.py`, `setup_script_path: scripts/setup.sh`, active state and yearly cron. Open the detail page and click **Run now**; correlate its `bash_command_id` with `api GET '/api/bash/bash_events/search?command_id__eq=<id>&limit=100'`. Expected: the configured nested setup runs before the entrypoint. Known failure, reproduced 2026-10-12 against automation 1.19.3: the real BashOutput exits 1 with the marker assertion, proving the setup step was skipped ([OpenHands/automation#343](https://github.com/OpenHands/automation/issues/343)). Only an owned marker is written; no external service or model is used.+- **Selected profile on a prompt preset (`F23.configuration`, `F23.run-linked-conversation`).** Arrange a bare saved Agent Profile (`tools: []`, unique persona) referencing an offline LLM profile whose base URL is the closed loopback `http://127.0.0.1:9`; set zero retries and disable title generation. Create an owned active annual-cron prompt preset, PATCH its `agent_profile_id` to the bare profile, then click **Run now** in Canvas. Inspect the linked conversation's `launched_agent_profile`, tool names and persona without revealing secrets. Expected: the chosen profile governs that conversation. Known failure, reproduced 2026-10-12 against automation 1.19.3/Agent Server 1.54.0: provenance is null, the selected persona is absent, and default terminal/file-editor/task-tracker/finish tools replace the bare selection ([OpenHands/automation#557](https://github.com/OpenHands/automation/issues/557)). The deliberate loopback connection failure is incidental to this metadata check; no provider is reachable.+## Gotchas- The automations QA checklist (F28) expects edit-mode **Test** to run the current form state; the panel calls `dispatchAutomation(<id>)`, which runs the saved automation. `F23.edit-test` records current behavior; whether it should change is a product decision, not a map correction.@@ -119,9 +124,9 @@ Preconditions:- The detail page has no `main` landmark: `browser snapshot 'role=main'` times out. Snapshot the page, or scope by section headings (`role=heading[name="Configuration"] >> xpath=ancestor::div[2]`).- The kebab menu has no test ids; `role=list >> has-text=Tarball` scopes the detail kebab (the dashboard kebab has no Tarball, so F21 uses `has-text=Export`).- Kebab **Edit** (detail and dashboard) opens `/automations/setup?automationId=<id>`. The edit page's Save and Test report API failures with the API's message (see `F23.edit-validation`, `F23.edit-save-error`), and its `Automation updated` toast covers Test/Save/Close until the pointer moves away.-- Known defect: clearing **At** on a preset schedule and saving stores midnight with no message (#17948 item; with the old modal it was silently ignored).+- Known defect: clearing **At** on a preset schedule and saving stores midnight with no message ([#17948](https://github.com/OpenHands/OpenHands/issues/17948) item; with the old modal it was silently ignored).- On a local backend the Logs dialog reads every run's output from the server-level `/api/bash/bash_events/search`, with or without a conversation. It caches what it fetched: reload before checking its requests with `browser network`.-- Known defect: prompt runs also have a `bash_command_id`, so every row shows the logs icon, and a prompt run that ended without a conversation is labelled as a script run (`F23.run-no-conversation`) (#17948).+- Known defect: prompt runs also have a `bash_command_id`, so every row shows the logs icon, and a prompt run that ended without a conversation is labelled as a script run (`F23.run-no-conversation`) ([#17948](https://github.com/OpenHands/OpenHands/issues/17948)).- The automation service (1.19.0 here) keeps repos and plugins only under `preset_metadata`; the UI reads them from there. Repositories show `owner/repo` without the host (a GitLab URL reads like a GitHub one), a repo without a ref gets no badge, and Export keeps only the first repository (all plugins). The **Notification** field reads a top-level `notification` that the service has no field for, so it never renders on a local stack (#16644 tracks removing or mapping it).- The Schedule field shows the raw cron (`0 9 * * *`) and no timezone: the service returns no `schedule_human` and no top-level `timezone`, so the Edit dialog also omits its `Timezone:` hint.- `run-phase` never appeared in the log here (`phase_code` stays `null` on the local service), so the phase chip is not reachable on this stack.@@ -130,8 +135,19 @@ Preconditions:- The script section needs a tarball automation. The account-free way to get one is F22's **Daily news digest** bundle entry (it uploads its bundle and creates the automation); the custom-automation **Upload tarball** action takes `control-openhands fixture tarball`. Its kebab Edit shows generated starter Python instead of the saved bundle (#18341, see F21.edit); a cold load of `/automations/setup?automationId=<id>` shows the **Existing code bundle** summary.- **Starting…** on Run now lasts only for the dispatch request (under 50 ms locally); `--observe` does not catch it.- The run row is a plain `<a href>`, so opening a run's conversation reloads the whole app (a `document` request in `browser network`), unlike sidebar navigation.-- A 404 detail request is retried three times before **Automation not found** shows (about 7 s of skeleton) (#17948); wait with `wait-text`, not `count`.+- A 404 detail request is retried three times before **Automation not found** shows (about 7 s of skeleton) ([#17948](https://github.com/OpenHands/OpenHands/issues/17948)); wait with `wait-text`, not `count`.- `browser fill` on the Timeout spinbutton with letters fails in Playwright and, after its 30 s retry, the dialog was found closed; use numeric values (`1.5`, `0`, `1801`) to reach the timeout errors.-- The Tarball download is named `<name>.tar` (spaces become `_`) but is gzip-compressed (#17948); `browser downloads --inspect` does not list tar entries, so inspect it with `tar -tzf`.+- The Tarball download is named `<name>.tar` (spaces become `_`) but is gzip-compressed ([#17948](https://github.com/OpenHands/OpenHands/issues/17948)); `browser downloads --inspect` does not list tar entries, so inspect it with `tar -tzf`.- **Debug with OpenHands** starts a real agent with the `openhands-automation` skill and the automation API key. It acts on live automations: here it looked up `QA_F23 Fail` by name (a newer fixture with the same name), PATCHed its timeout back to 600 and dispatched it, which adds runs and conversations. Stop it once it has started: on its page `control-openhands browser click 'testid=stop-button'`, then `control-openhands conversation wait <id> --until paused,finished,idle --timeout 45` (`paused`).- A successful run leaves a conversation `<name> — <UTC timestamp>` that stays after the automation is deleted (conversations family owns deleting it).+- Known issue [#15465](https://github.com/OpenHands/OpenHands/issues/15465) (`F23.configuration-event` `F23.activity-log`): An automation configured for Agent Canvas release events never runs automatically, although Run now works. Recheck a real signed GitHub/Forgejo release webhook through an externally reachable responder into a disposable automation and inspect its activity log; local manual dispatch cannot verify event delivery. No external repository webhook credentials or controlled release-event fixture are available.+- Known issue [OpenHands/software-agent-sdk#5279](https://github.com/OpenHands/software-agent-sdk/issues/5279) (`F23.run-linked-conversation`): An automation that attaches to an existing remote conversation can complete with no conversation link in its run, because the workspace callback does not register the attached id. Recheck a Cloud automation script that calls `RemoteConversation.attach` on an owned stable conversation and exits through the real workspace completion callback, then inspect the Activity Log link. Blocked without that authenticated Cloud workspace/callback context; a plain local GET is not callback proof.+- Known issue [OpenHands/software-agent-sdk#4705](https://github.com/OpenHands/software-agent-sdk/issues/4705) (`F23.run-status-polling`, `F23.run-logs-modal`): GitHub MCP tool events can omit `inputSchema`, causing an automation's RemoteConversation decoder/run log to report failure while the agent continues, so task and automation statuses disagree. Recheck a real GitHub-MCP-enabled run using the affected client/server versions and its retained tool-definition event; the issue does not give those versions. A generic MCP fixture does not establish the reported GitHub/version branch, and genuine run comparison needs a model key.+- Known issue [OpenHands/software-agent-sdk#4260](https://github.com/OpenHands/software-agent-sdk/issues/4260) (`F23.run-linked-conversation`, `F23.run-logs-modal`): Cloud PR-review automation can remain at review in progress and time out without posting a review; its event webhook also fails to serialize SecretStr, breaking the UI stream even on successful reviews. Keep the stream-serialization failure distinct from the empty-response/600-second run-budget failure. Recheck with an owned Cloud reviewer/webhook deployment, authorized disposable review target and real model credentials; a local script run proves neither production path.+- Known issue [OpenHands/automation#286](https://github.com/OpenHands/automation/issues/286) (`F23.run-status-polling`): Multi-worker/replica automation deployments can run duplicate watchdogs that query the same sandbox and hit rate limits, falsely failing runs. Recheck the real deployment's worker/replica configuration with an owned long-running Cloud run and sandbox verification traffic; this single-worker local stack cannot prove distributed scheduler/watchdog ownership.+- Known issue [OpenHands/automation#219](https://github.com/OpenHands/automation/issues/219) (`F23.run-status-polling`): Starting a Cloud automation at the user's sandbox cap can pause an existing interactive session instead of reaching the graceful Skipped–Limit Reached path. Recheck only on an owned disposable Cloud account at its recorded cap, with a live interactive fixture and a new automation run, then observe both run and conversation states. This run has no Cloud account/cap fixture; do not evict a production session as a test.+- Known issue [OpenHands/automation#113](https://github.com/OpenHands/automation/issues/113) (`F23.run-linked-conversation`): Scheduled plugin jobs can start conversations visible in the archived list while the automation run UI says no conversation started. Recheck the affected automation/run ids, archived conversation ids and deployment version, using the real callback relationship; these identifiers and version are not provided. This is a reported missing-link state, not proof that no agent ran.+- Known failure [OpenHands/automation#514](https://github.com/OpenHands/automation/issues/514) spans Home cancellation (`F03.cancel-in-flight-run`) and `F23.activity-log`/`F23.run-logs-modal`: on 2026-10-12 a genuine Home **Cancel run** changed the owned bounded local worker's row to **Cancelled**, while its Python entrypoint remained alive in the owned launcher ancestry. The F03 driving bullet owns upload/create/cancel steps and retained evidence. The worker later reached its 180 s bound and exited; a cancelled status alone does not prove process cleanup.+- Known issue [OpenHands/software-agent-sdk#4388](https://github.com/OpenHands/software-agent-sdk/issues/4388) (`F23.run-status-polling`, `F23.run-logs-modal`): incremental SDK Bash output polling can return empty pages before newer chunks when `order__gt` is applied after the file-page boundary. Canvas's frontend log reader does not send that filter, but automation launch preparation and prompt/plugin preset repository clones use `(Async)RemoteWorkspace.execute_command`, whose output poller does. Recheck a bounded genuine preset/setup or repository-clone command that emits enough chunks to cross the 100-event boundary, correlating its final BashOutput and real run result; a frontend Logs read or a one-output command does not cover the incremental branch.+- Known issue [OpenHands/software-agent-sdk#5937](https://github.com/OpenHands/software-agent-sdk/issues/5937) (`F23.run-status-polling`, `F23.run-logs-modal`): prompt/plugin preset workers attach `AUTOMATION_CONVERSATION_ID` for continued conversations and then call `RemoteConversation.run()`. If that same conversation is already running a tool for more than 30 s, the join request can time out and fail the automation worker while the agent continues. Recheck an owned continued-conversation automation with a genuine already-running long tool and correlate worker error, agent status and final run result; a fresh conversation or one normal first `run()` does not exercise the join. No model execution is authorized in this pass.+- Known issue [OpenHands/software-agent-sdk#5938](https://github.com/OpenHands/software-agent-sdk/issues/5938) (`F23.run-status-polling`): prompt/plugin preset workers register an event callback that forwards `ActionEvent.summary` as the live phase. After an events-socket reconnect, REST-reconciled events reach SDK state but are not replayed to that callback, so summaries emitted in the gap can be missed. Recheck an owned real preset worker through a bounded reconnect window while a genuine action is recorded; compare callback-driven phase with stored events. Current Canvas phase rendering has its separate [#17121](https://github.com/OpenHands/OpenHands/issues/17121) gap, and the worker's completion metadata is not claimed lost merely because this callback missed an event.diff --git a/.agents/skills/verify-openhands/references/feature-map/F24-git-sync.md b/.agents/skills/verify-openhands/references/feature-map/F24-git-sync.mdindex 11076562bc..f033973349 100644--- a/.agents/skills/verify-openhands/references/feature-map/F24-git-sync.md+++ b/.agents/skills/verify-openhands/references/feature-map/F24-git-sync.md@@ -93,7 +93,7 @@ Preconditions:- Feedback colours are semantic tokens (#18346): the check-failure block, the **Last sync error** banner and a `Sync failed` row use the danger surface and feedback-error text (`rgb(231, 106, 94)` on OpenHands-Neutral, `rgb(192, 40, 45)` on white under Light+), and `Sync complete` uses success (`rgb(165, 231, 94)` Neutral, `rgb(15, 115, 77)` Light+); verified 2026-10-13 on desktop. Read them with `browser eval` of `getComputedStyle(...).color` on the text elements inside `git-sync-check-failure`, `git-sync-error-banner` and `git-sync-activity-row` after switching themes on `/settings/app` (`F16.color-theme`).- Toasts (`Git Sync settings saved.`, `Enable Git Sync before triggering a sync.`) last a few seconds: read them with `browser wait-text` immediately after the click, never after a long `--observe` window.- Save's label cycles `Save Changes` → `Checking repository...` (only when a checked field changed) → `Saving...` → `Save Changes` in about 100 ms against a local repo; `--observe 'testid=git-sync-save-button'` records it.-- Setting, rotating or clearing the encryption key re-exports the automation even when its definition was unchanged. Verified 2026-10-10 on pinned automation `1.19.3`, Agent Server `1.54.0`, desktop and phone, selected backend Local: each key-only save sets `dirty_count` to `1`; the next sync creates a commit with ciphertext after set, different ciphertext after rotation and plaintext after clear ([OpenHands/automation#563](https://github.com/OpenHands/automation/pull/563), fixing [#551](https://github.com/OpenHands/automation/issues/551)). Read the dirty count and exported file before any automation PATCH; a separate dirty export tests a different contract.+- Setting, rotating or clearing the encryption key re-exports the automation even when its definition is unchanged. Each key-only save sets `dirty_count` to `1`; the next sync creates a commit with ciphertext after set, different ciphertext after rotation and plaintext after clear. Read the dirty count and exported file before any automation PATCH; a separate dirty export tests a different contract.- Nothing on the page says whether a token is stored: the token placeholder is always `Leave blank to keep the current token`, and the author fields always reload blank. Prove author changes from the commit (`git log --format='%an <%ae>'`).- The card's Repository value drops embedded credentials, as link text and as plain text: an http(s) URL loses its whole `user:token@` (a token alone in the user slot too), other schemes keep the user and lose only the password (`ssh://git@host:2222/...`), and the rest stays as configured. The Repository URL input and `GET /api/automation/v1/git-sync/status` still carry the credentials, so a screenshot of the form shows them. Never type a real token into the URL.- The page has no Automate sub-navigation (Dashboard / Templates) and no command-menu entry; leave through **Back to Automations** (`role=link`, not a button).@@ -101,3 +101,4 @@ Preconditions:- An untouched **Enable Git Sync** switch follows the server: after another admin pauses sync, the next idle poll flips both the pill and the switch off, so turning it back on is one label click plus Save.- `F24.unsupported` uses `--automation-ref` for the old Git tag; without access to that source it stays blocked. The current CLI also supports `--automation-version V` for a PyPI release (verified with 1.19.3 on 2026-10-10). Backend overrides come from these explicit launch flags; shell `OH_AUTOMATION_VERSION` is not forwarded.- Arrange with `api ... --write` only to mark automations changed or to simulate another admin; the save, check and sync steps must go through the form.+- Known issue [OpenHands/automation#586](https://github.com/OpenHands/automation/issues/586) (`F24.sync-now`): a YAML-only repository edit can replace an unchanged uploaded script bundle because canonical gzip bytes retain their creation second. Recheck a disposable Git Sync repository and script automation, change only YAML, arrange the canonical rebuilds across different seconds, and compare the real saved upload reference before and after sync. Equal decompressed contents or a serializer-only probe do not prove that Canvas reuses its upload.diff --git a/.agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md b/.agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.mdindex c52c4c24ec..4068e4f4ca 100644--- a/.agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md+++ b/.agents/skills/verify-openhands/references/feature-map/F25-backends-and-cloud.md@@ -79,7 +79,7 @@ Preconditions:- Cloud account: where this machine's proxy injects OpenHands Cloud credentials (`curl -s -o /dev/null -w '%{http_code}' https://app.all-hands.dev/api/organizations` is `200` without a token), the **Cloud backend** bullet below can approve the device flow in its popup; otherwise it is blocked. Blocked here even with such an account: `F25.cloud-log-back-in` (needs a logged-out Cloud row), org switching in `F25.cloud-org-rows` (needs a second organization), the archived and error states of `F25.cloud-sandbox-states`, `F25.cloud-org-suspended` (also needs a suspended OpenHands Enterprise organization or membership), `F25.share-publicly` (toggle), and `F25.locked-cloud`, which also needs a launcher flag (see Gotchas).- **Open the selector (`F25.selector-dropdown`).** From `/conversations` run `control-openhands browser click 'testid=backend-selector'` and `control-openhands browser snapshot 'testid=backend-selector'`. The snapshot shows `listbox` with `option "Connected Local"`, a separator, `button "Add Backend"` and `button "Manage Backends"`. Screenshot with `control-openhands browser screenshot --feature F25.selector-dropdown --name open`. Hover opens it too: `browser press Escape`, move the pointer away with `control-openhands browser hover 'testid=command-menu-trigger'` (hovering a selector the pointer is already on does nothing), `control-openhands browser hover 'testid=backend-selector'`, then the same snapshot shows `combobox "Local" [expanded]` and the two footer buttons.-- **Add Backend modal (`F25.add-backend-modal`).** Run `control-openhands browser click 'testid=add-backend-menu-item'` and `control-openhands browser snapshot 'testid=add-backend-modal'`: heading `Choose how you want to connect`, link `Compare Cloud, self-hosted, and Enterprise deployment options` (`/url: https://docs.openhands.dev/overview/introduction`, text `Learn more`) and tab `OpenHands Cloud ...` `[selected]` next to tab `Agent-server ...`. Keyboard focus: `control-openhands browser eval "(()=>{const a=document.activeElement;const d=document.querySelector('[data-testid=add-backend-modal]');return {tid:a.getAttribute('data-testid'),inDialog:!!(d&&d.contains(a))}})()"` is `{tid: "add-backend-close", inDialog: true}` right after opening; after eight `control-openhands browser press Tab` and three `browser press Shift+Tab` (reading the same eval after each) `inDialog` stays `true` (Tab past **Advanced** wraps to the close button), and `browser press q` then `control-openhands browser eval "document.querySelector('[data-testid=chat-input]').textContent"` is `""` (nothing reaches the Home composer behind). `browser screenshot --feature F25.add-backend-modal --name focus-trap-open`. `control-openhands browser press Escape` leaves `browser count 'testid=add-backend-modal'` at `1`; `control-openhands browser click 'testid=add-backend-close'` makes it `0`, and `control-openhands browser eval "!!document.activeElement.closest('[data-testid=backend-selector]')"` is `true` (focus is back on the selector). Keyboard focus in a light theme: arrange `control-openhands browser eval "localStorage.setItem('openhands-color-theme','light-plus')"` and `browser reload`, reopen the modal, run `control-openhands browser focus 'testid=add-backend-option-cloud'` and `browser press ArrowRight`, then `control-openhands browser eval "getComputedStyle(document.activeElement).outlineColor"`. Expected: the theme's focus colour (`--oh-focus`, `#007ACC` in Light+). Known failure (reproduced 2026-10-13, #18324): a fixed pale `outline-blue-300` (`oklab(0.79 -0.03 -0.10)`), about 1.8:1 on the light tab (`browser screenshot 'testid=add-backend-modal' --feature F25.add-backend-modal --name tab-focus-light-plus`). Restore with `browser eval "localStorage.removeItem('openhands-color-theme')"`.+- **Add Backend modal (`F25.add-backend-modal`).** Run `control-openhands browser click 'testid=add-backend-menu-item'` and `control-openhands browser snapshot 'testid=add-backend-modal'`: heading `Choose how you want to connect`, link `Compare Cloud, self-hosted, and Enterprise deployment options` (`/url: https://docs.openhands.dev/overview/introduction`, text `Learn more`) and tab `OpenHands Cloud ...` `[selected]` next to tab `Agent-server ...`. Keyboard focus: `control-openhands browser eval "(()=>{const a=document.activeElement;const d=document.querySelector('[data-testid=add-backend-modal]');return {tid:a.getAttribute('data-testid'),inDialog:!!(d&&d.contains(a))}})()"` is `{tid: "add-backend-close", inDialog: true}` right after opening; after eight `control-openhands browser press Tab` and three `browser press Shift+Tab` (reading the same eval after each) `inDialog` stays `true` (Tab past **Advanced** wraps to the close button), and `browser press q` then `control-openhands browser eval "document.querySelector('[data-testid=chat-input]').textContent"` is `""` (nothing reaches the Home composer behind). `browser screenshot --feature F25.add-backend-modal --name focus-trap-open`. `control-openhands browser press Escape` leaves `browser count 'testid=add-backend-modal'` at `1`; `control-openhands browser click 'testid=add-backend-close'` makes it `0`, and `control-openhands browser eval "!!document.activeElement.closest('[data-testid=backend-selector]')"` is `true` (focus is back on the selector). Keyboard focus in a light theme: arrange `control-openhands browser eval "localStorage.setItem('openhands-color-theme','light-plus')"` and `browser reload`, reopen the modal, run `control-openhands browser focus 'testid=add-backend-option-cloud'` and `browser press ArrowRight`, then `control-openhands browser eval "getComputedStyle(document.activeElement).outlineColor"`. Expected: the theme's focus colour (`--oh-focus`, `#007ACC` in Light+). Connection tabs, the guidance toggle and the DigitalOcean provider choice use the theme focus role and preserve host `--oh-focus` overrides; default focus contrast is at least 3:1 against base, modal, raised and hovered neutral surfaces in all five themes. Solarized Light uses `#227FBE`. Check the settled outline after its transition, preserving the 2 px width and existing offsets. Restore with `browser eval "localStorage.removeItem('openhands-color-theme')"`.- **Advanced host (`F25.cloud-advanced-host`).** Reopen with `browser click 'testid=backend-selector'` and `browser click 'testid=add-backend-menu-item'`, run `control-openhands browser click 'testid=add-backend-advanced-toggle'`, then `control-openhands browser attr 'testid=add-backend-advanced-toggle' aria-expanded` (`true`) and `control-openhands browser snapshot 'testid=add-backend-cloud-panel'` (textbox `Host` with placeholder `https://app.all-hands.dev`).- **Device flow error (`F25.cloud-device-flow`).** Run `control-openhands browser fill 'testid=add-backend-cloud-host' http://127.0.0.1:9`, `control-openhands browser click 'testid=add-backend-login-button' --observe 'testid=add-backend-device-flow'` and `control-openhands browser wait 'testid=add-backend-auth-error' --timeout 15000`. The observation goes `Starting authentication...` → `Failed to start device flow: Failed to fetch` / `Try again`, and `control-openhands browser network --last 5` shows `POST /oauth/device/authorize` to `http://127.0.0.1:9`. `control-openhands browser click 'testid=add-backend-auth-retry'` starts again and fails the same way. Each attempt leaves an `about:blank` popup: `control-openhands browser tabs`, then `control-openhands browser close-tab <i>` for each extra page and `control-openhands browser tab 0`. Close the modal with `testid=add-backend-close`.- **Device flow awaiting and Cancel (`F25.cloud-device-flow`).** Reopen Add Backend (Cloud tab, default host), run `control-openhands browser click 'testid=add-backend-login-button' --observe 'testid=add-backend-device-flow' --observe-ms 6000`, then `control-openhands browser snapshot 'testid=add-backend-device-flow'`. If `https://app.all-hands.dev` is reachable (`control-openhands browser network --external --last 5` lists it), the snapshot shows `Waiting for authorization...`, `Browser opened. Complete sign-in to continue. If browser didn't open, visit:`, a link `https://app.all-hands.dev/oauth/device/verify?user_code=<code>` and button `Cancel`; `browser tabs` lists a second page on that URL. Screenshot with `--feature F25.cloud-device-flow --name awaiting`, then `control-openhands browser click 'testid=add-backend-auth-cancel'`: the panel is back to `Connect to OpenHands` / `Advanced` and `browser tabs` lists one page. Do not approve the code here (the **Cloud backend** bullet does, when it can). If Cloud is unreachable you get the error state of the previous bullet instead.@@ -110,6 +110,11 @@ Preconditions:- **Cleanup.** `browser press Escape`, then `control-openhands stop --run "$OH_VERIFY_RUN_2"` and `unset OH_VERIFY_RUN_2`. The conversation stays with this run's state.- **Errors.** `control-openhands browser errors --app-only` after the family shows no page errors. Expected app-origin entries: `POST /oauth/device/verify-authenticated` 404 (twice), the shared-view 404s, the warning `Scripts may close only the windows that were opened by them.` from the device-verify Cancel, and a burst of CORS `console.error`s for `<second-url>` while the second agent server is stopped or restarting (see Gotchas).+- **Same-origin credential readability (`F25.manage-backends`, `F25.add-agent-server`).** In an authenticated private run, use read-only `browser eval` to return only the count of registered entries and the count whose `apiKey` is a nonempty string in `JSON.parse(localStorage.getItem('openhands-backends') || '[]')`. Expected: backend credentials are not readable by arbitrary same-origin script. Known failure, reproduced 2026-10-12: registered entries have readable plaintext keys ([#16492](https://github.com/OpenHands/OpenHands/issues/16492)). Retain counts/booleans only, never key values or whole storage JSON.+- **Manually registered same-origin key rotation (`F25.add-agent-server`, `F25.recovery-gate`).** Through **Add Backend → Agent-server**, add an owned entry named `QA_Rotation_Legacy` with the run's exact origin and fill its API Key using `--value-file <run>/private/session-key`; keep it active. Run `restart --rotate-key`, `doctor`, and `browser reload`. Compare only each saved entry's `apiKey === window.__AGENT_CANVAS_SESSION_API_KEY__` boolean. Expected: same-origin entries accept the newly injected session key and the active backend recovers. Known failure, reproduced 2026-10-12: `default-local` matches the injected key, the manually registered entry does not, and the UI shows **Disconnected**/**Unauthorized** ([#17371](https://github.com/OpenHands/OpenHands/issues/17371)). This version has visible status feedback, so do not repeat the older claim of a completely silent loop. Choose **Connected Local** to recover, then remove the owned duplicate.++- **Non-loopback default backend name (`F25.selector-dropdown`).** Serve the unchanged frontend on an owned non-loopback LAN address with `--auth-required`, no session-key injection and real protected backend routes. Open that origin in a new browser tab with `--allow-external`; its Add a backend step defaults **Host Name** to `Local`. Authenticate only with the private run key file, continue once and skip the remaining setup, then inspect `backend-selector`. Expected: the default name identifies the backend's host on a remote origin. Known failure, reproduced 2026-10-12: the selector reads **Connected Local** on the non-loopback origin ([#16297](https://github.com/OpenHands/OpenHands/issues/16297)). This is a LAN-origin check, not a VM deployment claim. Stop only the owned fixture frontend and return to the original tab.+## Gotchas- Export the second run as `OH_VERIFY_RUN_2` and pass it with `--run`; never export it as `OH_VERIFY_RUN`, or the browser commands look for a browser that run does not have. `OH_VERIFY_RUN= control-openhands launch --new ...` keeps the launch from reusing this run.@@ -123,10 +128,10 @@ Preconditions:- A tab opened with `click --modifiers Control` starts with an empty sessionStorage, so it boots from the localStorage fallback (the backend last chosen in any tab) unless its URL carries `?backend=`. Opening a `?backend=` link writes that backend to localStorage too: a per-tab check that relies on "last choice" must run before opening pinned links.- The Add Backend Agent-server form has no inline "required" messages; it only disables **Connect**. The messages exist in the Edit form.- Footer clicks from the collapsed popover or the phone drawer open the modals a moment later: `browser wait` for them instead of `count`.-- The device-flow popup survives errors: every failed attempt leaves an `about:blank` tab (#17953). Cancel closes it; after an error close it with `browser close-tab`.+- The device-flow popup survives errors: every failed attempt leaves an `about:blank` tab ([#17953](https://github.com/OpenHands/OpenHands/issues/17953)). Cancel closes it; after an error close it with `browser close-tab`.- With the default host the device flow talks to the real `app.all-hands.dev` (`POST /oauth/device/authorize`, then token polling). Starting and cancelling is harmless; approve a code only in the **Cloud backend** bullet, with an account you may use.- While the second stack's agent server is stopped or restarting, its ingress answers without CORS headers, so the browser logs `Access to fetch at '<second-url>/api/settings' ... blocked by CORS policy` (also `/server_info`, `/api/conversations/search`, `/api/llm/models/verified`). That is the dead upstream, not a CORS bug; a healthy second stack sends `access-control-allow-origin` for this origin.-- Expected: Escape never clears the backend selector. Known failure (reproduced 2026-10-09, 2026-10-12, 2026-10-13 and 2026-10-14, #18233): Escape on the focused backend selector while its menu is closed (a second Escape after the one that closes the menu, and sometimes the Escape that closes Manage Backends opened from it) clears the selection: the closed selector shows the active backend's name as an italic placeholder (`browser value 'testid=backend-selector >> role=combobox'` is `""`) and its status dot is gone, so `browser wait 'testid=backend-selector >> role=status[name="Disconnected"]'` times out although the backend is down; a reload restores it. Reload before a closed-selector status check, as the Edit bullet does. Repro: `browser click 'testid=backend-selector'`, `browser press Escape` twice (#18233). On 2026-10-12 `bug proof` verified fix PR #18238 at 77de0ad24 against main f07aebdf5 (two `xfail`, two `xpass`, desktop). `control-openhands bug run F25.selector-dropdown` replays it (desktop, after a reload with the mouse off the selector: a reload under a mouse left on it opens its menu and empties its field, which would look like the bug) and reports `xfail` while the bug lasts:+- Expected: Escape never clears the backend selector. Known failure (reproduced 2026-10-09, 2026-10-12, 2026-10-13 and 2026-10-14, [#18233](https://github.com/OpenHands/OpenHands/issues/18233)): Escape on the focused backend selector while its menu is closed (a second Escape after the one that closes the menu, and sometimes the Escape that closes Manage Backends opened from it) clears the selection: the closed selector shows the active backend's name as an italic placeholder (`browser value 'testid=backend-selector >> role=combobox'` is `""`) and its status dot is gone, so `browser wait 'testid=backend-selector >> role=status[name="Disconnected"]'` times out although the backend is down; a reload restores it. Reload before a closed-selector status check, as the Edit bullet does. Repro: `browser click 'testid=backend-selector'`, `browser press Escape` twice ([#18233](https://github.com/OpenHands/OpenHands/issues/18233)). On 2026-10-12 `bug proof` verified fix PR #18238 at 77de0ad24 against main f07aebdf5 (two `xfail`, two `xpass`, desktop). `control-openhands bug run F25.selector-dropdown` replays it (desktop, after a reload with the mouse off the selector: a reload under a mouse left on it opens its menu and empties its field, which would look like the bug) and reports `xfail` while the bug lasts:```sh known-bug F25.selector-dropdown escape-closedcontrol-openhands browser viewport desktop@@ -143,11 +148,19 @@ Preconditions:```- Local and remote backends are probed every 30 s (settings + server_info), Cloud backends every 5 min (#18128), so allow up to 40 s for a local dot to change. `browser network --clear`, wait about 70 s, then `browser network --filter server_info` shows two probes per local origin 30 s apart. While the active backend is down, any reload lands on the recovery gate: drive the disconnected rows on the page that was already loaded.- `/oauth/device/verify` is a Cloud page that also ships in Canvas: on a local stack its POST target does not exist, so Authorize always ends in `Error`. The "Authentication Required" state is unreachable (`useIsAuthed` always resolves `true`).-- `/shared/conversations/<id>` on a local backend also raises a raw toast `HTTP request failed (404 Not Found): {"detail":"Not Found"}` next to `Conversation not found` (#17953).-- Known failure (repro candidate, not an exemption): `F25.edit-escape` closes the Manage modal underneath and drops the Edit draft (Cancel and X close only Edit, `F25.edit-cancel`) (#17953).-- The add bullet's `network --external --filter 'ids='` check guards an earlier race: on a backend's first activation from `/conversations/<id>`, about 1 attempt in 5 asked the new backend for the old conversation id. It did not happen in 15 first activations at 4eff80237: each time `QA_Second` was removed, its consent unset with `control-openhands api PATCH /api/settings --run "$OH_VERIFY_RUN_2" --write --data '{"misc_settings_diff":{"app_preferences":{"user_consents_to_analytics":null}}}'` (arrange only) and the backend added again from `/conversations/<id>`. A backend that lacks an id answers `[null]`, which now reads as "not found" (only the `This conversation does not exist…` toast), not as incompatible data. No merged change was shown to fix the race itself (#18009 only changed how a `[null]` answer reads; #18024 only moved the redirect into a shared hook, still called after `addBackend`), so a hit (an `ids=` request to `<second-url>`) is a bug to file. At 95115e84f6dd it hit on 1 of 2 first activations from `/conversations/<id>` (one `GET <second-url>/api/conversations?ids=<id>` answered `200`, with no `This conversation does not exist…` toast after it). The hit followed a failed Connect in the same modal (an empty key, `Invalid API key`) before the successful one; the clean retry, which connected once, had none, so a repro should try both sequences. On 2026-10-08 at d2c89252d it did not reproduce in 17 first activations (8 clean, 9 failed-then-clean, some under CPU load and some human-paced): every `ids=` capture was empty, and no merged change touched the activation logic since the hit, so it stays a watch item rather than a filed defect. Record the Add bullet as `fail` when the request appears, even though the redirect and the selector look right. Known failure (reproduced 2026-10-13, #18390): two independent first activations hit it that day, each the run's first activation of `QA_Second` after a clean Connect. On 2026-10-13 at 2bdc0e9d6 it hit on 1 of 6 first activations from `/conversations/<id>`: the first one of the run (`GET <second-url>/api/conversations?ids=<id>` `200`, no toast), a clean Connect in a modal opened after earlier failed Connects in other Add Backend openings; the next 5 (2 failed-then-clean, 3 clean, plus one Manage → Add) had none.-- Add Backend and Manage Backends take and keep keyboard focus, and Edit backend takes it on open (`ModalBackdrop`; at phone width open the selector with `'testid=sidebar-mobile-drawer >> testid=backend-selector'`, since the bare test id matches two selectors). At desktop width focus returns to the selector input when they close (at phone width Add Backend's X closes the drawer too and focus falls to the page body), so an Escape pressed after closing lands on the selector and can trigger #18233 below.+- `/shared/conversations/<id>` on a local backend also raises a raw toast `HTTP request failed (404 Not Found): {"detail":"Not Found"}` next to `Conversation not found` ([#17953](https://github.com/OpenHands/OpenHands/issues/17953)).+- Known failure (repro candidate, not an exemption): `F25.edit-escape` closes the Manage modal underneath and drops the Edit draft (Cancel and X close only Edit, `F25.edit-cancel`) ([#17953](https://github.com/OpenHands/OpenHands/issues/17953)).+- The add bullet's `network --external --filter 'ids='` check guards an earlier race: [#18390](https://github.com/OpenHands/OpenHands/issues/18390) tracks the same first-activation symptom; on a backend's first activation from `/conversations/<id>`, about 1 attempt in 5 asked the new backend for the old conversation id. It did not happen in 15 first activations at 4eff80237: each time `QA_Second` was removed, its consent unset with `control-openhands api PATCH /api/settings --run "$OH_VERIFY_RUN_2" --write --data '{"misc_settings_diff":{"app_preferences":{"user_consents_to_analytics":null}}}'` (arrange only) and the backend added again from `/conversations/<id>`. A backend that lacks an id answers `[null]`, which now reads as "not found" (only the `This conversation does not exist…` toast), not as incompatible data. No merged change was shown to fix the race itself (#18009 only changed how a `[null]` answer reads; #18024 only moved the redirect into a shared hook, still called after `addBackend`), so a hit (an `ids=` request to `<second-url>`) is a bug to file. At 95115e84f6dd it hit on 1 of 2 first activations from `/conversations/<id>` (one `GET <second-url>/api/conversations?ids=<id>` answered `200`, with no `This conversation does not exist…` toast after it). The hit followed a failed Connect in the same modal (an empty key, `Invalid API key`) before the successful one; the clean retry, which connected once, had none, so a repro should try both sequences. On 2026-10-08 at d2c89252d it did not reproduce in 17 first activations (8 clean, 9 failed-then-clean, some under CPU load and some human-paced): every `ids=` capture was empty, and no merged change touched the activation logic since the hit, so it stays a watch item rather than a filed defect. Record the Add bullet as `fail` when the request appears, even though the redirect and the selector look right. Known failure (reproduced 2026-10-13, #18390): two independent first activations hit it that day, each the run's first activation of `QA_Second` after a clean Connect. On 2026-10-13 at 2bdc0e9d6 it hit on 1 of 6 first activations from `/conversations/<id>`: the first one of the run (`GET <second-url>/api/conversations?ids=<id>` `200`, no toast), a clean Connect in a modal opened after earlier failed Connects in other Add Backend openings; the next 5 (2 failed-then-clean, 3 clean, plus one Manage → Add) had none.+- Add Backend and Manage Backends take and keep keyboard focus, and Edit backend takes it on open (`ModalBackdrop`; at phone width open the selector with `'testid=sidebar-mobile-drawer >> testid=backend-selector'`, since the bare test id matches two selectors). At desktop width focus returns to the selector input when they close (at phone width Add Backend's X closes the drawer too and focus falls to the page body), so an Escape pressed after closing lands on the selector and can trigger [#18233](https://github.com/OpenHands/OpenHands/issues/18233) below.- Toasts are not page or HTTP errors, so `browser errors --app-only` misses them: read `browser toasts --history` after every add or switch, before anything reloads the page (the history starts with each page load).- `F25.locked-cloud` needs `scripts/static-server.mjs --lock-to-cloud <url>`, which neither `bin/agent-canvas.mjs` nor `control-openhands launch` can pass today; even with it, reconnecting needs a Cloud account.- The Cloud tab title truncates to `OpenHan…` at 390 px when the Agent-server tab is selected; the layout still fits. With the Providers tab present all three titles are cut to three letters.- A token DigitalOcean rejects surfaces as the bare `401 Unauthorized`, with no hint that the token is wrong or lacks scopes.+- Known issue [#18029](https://github.com/OpenHands/OpenHands/issues/18029) (`F25.locked-cloud`): A returning user in a locked Cloud deployment can reach the no-backend placeholder and fail to start conversations despite a healthy configured backend. The report proposes a stale document or alternate origin but does not capture which occurred. Recheck against the affected locked deployment with a Cloud session, its exact origin, cache response headers and injected host state; the normal local launcher does not reproduce this deployment condition.+- Known issue [#17390](https://github.com/OpenHands/OpenHands/issues/17390) (`F25.cloud-sandbox-states`): Every sent Cloud message can fail with `token_not_found_in_db` from the model proxy. Recheck with the affected Cloud account/session and a minimal no-tools prompt; capture the error type with provider token text and hashes redacted. The report supplies no deployment version or deterministic token-provisioning sequence, and this run has no Cloud account.+- Known issue [#17163](https://github.com/OpenHands/OpenHands/issues/17163) (`F25.cloud-sandbox-states`): A Free Cloud account can lose access to previously usable models with HTTP 403 and a team-restriction message despite no team being created. Recheck with the affected account and its model entitlement/org history; local credentials cannot prove Cloud eligibility and this run has no such account.+- Known issue [#16874](https://github.com/OpenHands/OpenHands/issues/16874) (`F25.cloud-sandbox-states`): A failed resume of a paused Cloud sandbox was reported to consume a one-shot guard and leave chat disconnected until remount. Current source uses a retrying resume hook, so recheck the original failure and a second pause in one route mount before concluding resolution. Blocked without a real paused Cloud sandbox and a controlled one-request backend/network failure; source changes alone are not a live pass.+- Known issue [#15676](https://github.com/OpenHands/OpenHands/issues/15676) (`F25.cloud-sandbox-states`): Refreshing an approximately 24-hour-old Cloud conversation on Vercel can fail to resume/reconnect. Recheck an authenticated affected Cloud deployment with an aged sandbox/session and exact console/network error; the report omits its commit and failure text, and no aged Cloud fixture is available here.+- Known issue [#15822](https://github.com/OpenHands/OpenHands/issues/15822) (`F25.switch-backend` `F25.health-status` `F25.cloud-org-rows`): Cloud-proxy traffic was reported to continue after switching from Cloud to Local. Current source probes registered inactive Cloud backends on a five-minute cadence; recheck the reported repeated OPTIONS behavior with a real registered Cloud account/org and capture only credential-redacted request metadata after switching. This run has no authenticated Cloud backend.+- Known issue [OpenHands/automation#565](https://github.com/OpenHands/automation/issues/565) (`F25.cloud-org-rows`, `F25.cloud-org-suspended`): An org-scoped Cloud automation request with an unauthorized `X-Org-Id` can turn the identity service's 403 into a misleading 502/outage page. Recheck with an authenticated Cloud org-bound key and a disposable unauthorized organization id, comparing automation list and the identity response while preserving credential redaction. Local mode bypasses this identity service and cannot reproduce it.+- Known issue [OpenHands/software-agent-sdk#5789](https://github.com/OpenHands/software-agent-sdk/issues/5789) (`F25.add-agent-server`, `F25.health-status`): a deferred-init Agent Server can report ready after accepting a GUI origin through `/api/init`, while browser preflights from that origin still fail and Canvas cannot connect. Recheck an owned warm-pool server initialized with the actual non-localhost Canvas origin and compare admission with an origin supplied at server boot. The default loopback-origin stack is already allowed and cannot exercise this branch.diff --git a/.agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md b/.agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.mdindex f101764527..dfd5e12f7b 100644--- a/.agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md+++ b/.agents/skills/verify-openhands/references/feature-map/F26-runtime-variants.md@@ -26,13 +26,13 @@ Source: `bin/agent-canvas.mjs`, `scripts/dev-with-automation.mjs`, `scripts/dev-- `F26.session-key-persist`: in local mode without `LOCAL_BACKEND_API_KEY` the launcher generates a session key, saves it under `$HOME/.openhands/agent-canvas/api-key.txt` and injects the same key again on the next start.- `F26.session-key-pinned`: `LOCAL_BACKEND_API_KEY=<key>` in local mode injects exactly that key into the page and the API accepts only it (the saved generated key gets 401).- `F26.frontend-only`: `--frontend-only` serves the SPA; `/server_info`, `/api/*`, `/sockets/*` and `/api/automation/*` answer 503, and a browser new to that origin opens on the **Add a backend** onboarding step without an error toast.-- `F26.frontend-only-returning`: on a frontend-only origin, a browser that already stores a backend for that origin (the `default-local` entry a full launcher on the same port seeded, or one added there earlier) gets the recovery gate instead of onboarding: `agent-server-onboarding-screen` with Manage backends in recovery mode and the `Local` row `Disconnected` with the 503 detail (the gate itself is F25's recovery-gate row); the failing probes raise generic `An error occurred` toasts there (Known failure, reproduced 2026-10-08: #18160).+- `F26.frontend-only-returning`: on a frontend-only origin, a browser that already stores a backend for that origin (the `default-local` entry a full launcher on the same port seeded, or one added there earlier) gets the recovery gate instead of onboarding: `agent-server-onboarding-screen` with Manage backends in recovery mode and the `Local` row `Disconnected` with the 503 detail (the gate itself is F25's recovery-gate row); the failing probes raise generic `An error occurred` toasts there (Known failure, reproduced 2026-10-08: [#18160](https://github.com/OpenHands/OpenHands/issues/18160)).- `F26.backend-only`: `--backend-only` serves the APIs (key-protected) and answers 503 `No backend configured for this route` for `/` and static assets.- `F26.cross-connect`: a frontend-only UI connects to a separate backend-only instance through **Add a backend**; the shell then shows that backend as Connected and all API traffic goes to it.- `F26.remote-backend`: from a normal Canvas, backend selector → **Add Backend** → **Agent-server** → **Remote** shows the self-hosting guidance and connects to another instance (stand-in for a self-hosted VM, `docs/SELF_HOSTING.md`).- `F26.seeded-local-backend`: the first load of a launcher-served UI seeds one backend `Local` (id `default-local`) with the page origin and the injected key.- `F26.host-bind-lan`: `--host 0.0.0.0` warns, listens on the machine's LAN address, and serves an index.html that carries `__AGENT_CANVAS_AUTH_REQUIRED__` and neither `__AGENT_CANVAS_SESSION_API_KEY__` nor the key itself (the loopback launcher's page carries the key marker); the UI behaves like `--public` (Add a backend with Next disabled; Skip leads to the API-key screen).-- `F26.host-bind-lan-optin`: the opt-in the launcher's warning recommends (`--allow-lan-session-key`) should restore key injection on a LAN bind. Known failure (reproduced 2026-10-08): the npm launcher ignores the flag and the page still carries only `__AGENT_CANVAS_AUTH_REQUIRED__`; #17949.+- `F26.host-bind-lan-optin`: the opt-in the launcher's warning recommends (`--allow-lan-session-key`) should restore key injection on a LAN bind. Known failure (reproduced 2026-10-08): the npm launcher ignores the flag and the page still carries only `__AGENT_CANVAS_AUTH_REQUIRED__`; [#17949](https://github.com/OpenHands/OpenHands/issues/17949).- `F26.host-bind-env`: `OH_BIND_HOST=0.0.0.0` (the environment form of `--host`) prints the same two "not loopback" warnings and does not inject the session key, even with `LOCAL_BACKEND_API_KEY` set.- `F26.automation-startup-failure`: when the requested automation backend exits before it is ready, the launcher should print no success banner, stop the services it started and exit nonzero with the automation's exit reason. Known failure (reproduced 2026-10-13, #18362): it prints the **Main UI** banner and keeps the stack up with `/api/automation/health` answering `502`.- `F26.mars-web-bridge`: every launcher-served web stack hosts the DigitalOcean Managed Agents bridge on its ingress by default (`MARS_WEB=0` opts out): `/mars/health` answers `{"ok":true,"authRequired":true}` on a full, `--backend-only`, `--frontend-only` and `--host 0.0.0.0` launcher, `/mars/rpc/*` answers `401` without the session key and works with it, the static child behind the ingress hosts none, and the Electron stack hosts none either (its preload provides `window.marsBridge`).@@ -102,6 +102,14 @@ Preconditions:- **Desktop splash and window (`F26.desktop-boot-splash`, `F26.desktop-main-window`).** Start a private display and the app in one process group: `Q=$OH_VERIFY_RUN/private/f26-desktop; mkdir -p $Q/shots; setsid nohup sh -c "echo \$\$ > $Q/pgid; Xvfb :78 -screen 0 1440x1000x24 -nolisten tcp & sleep 1; DISPLAY=:78 HOME=$Q/home OH_CANVAS_SAFE_STATE_DIR=$Q/state LOCAL_BACKEND_API_KEY=qa-f26-desktop-key OH_SECRET_KEY=qa-f26-secret DO_NOT_TRACK=1 exec node_modules/.bin/electron --no-sandbox electron" > $Q/desktop.log 2>&1 &`, then capture frames: `for i in $(seq 1 40); do sleep 1; DISPLAY=:78 import -window root $Q/shots/s$(printf %02d $i).png; done`. Read the frames: the first seconds show the splash (logo, `OpenHands Agent Canvas`, `AI coding agent interface`, spinner, `Starting backend services…`, then lines such as `agent-server: {"asctime": ...`, the note `First launch downloads Python + the OpenHands agent server.` and a **Show details** button); a later frame shows the main window with a `File Edit View Window` menu and the first-run **Choose your agent** step under the consent dialog naming `the local backend "Local" at http://localhost:8000` (about 30 s on a warm cache, minutes on the first download). While it runs, `curl -s http://127.0.0.1:8000/mars/health | head -c 15` prints `<!DOCTYPE html>`: the desktop stack's ingress hosts no bridge (`marsWebBridge: false`, `F26.mars-web-bridge`). Copy the frames you cite into `$OH_VERIFY_RUN/evidence/F26.desktop-boot-splash/` and `.../F26.desktop-main-window/`. Stop in two steps: `kill -TERM $(cat $Q/pgid)` signals only the Electron launcher (it quits and stops its backend services); after about 10 s ports 8000, 18000, 18001 and 3001 are free again. Then `kill -TERM -- -$(cat $Q/pgid)` ends Xvfb; `ps -o pid,cmd -g $(cat $Q/pgid)` lists nothing.- **External links (`F26.desktop-external-links`).** Not driven: record `not-run` (harness gap: no verb clicks inside the Electron window or observes the OS browser handoff).+- **KV survives authentication rotation (`F26.session-key-rotated`, `F26.session-key-persist`).** With no explicit `AUTOMATION_KV_SECRET`, create an owned inert automation and PUT a harmless marker to `/api/automation/v1/kv/qa-session-rotation?automation_id=<id>` through `api --write`; GET must return that marker. Run `restart --rotate-key` and `doctor`, then GET again using the helper's fresh private key. Expected: rotating authentication preserves encrypted automation state. Known failure, reproduced 2026-10-12 against automation 1.19.3: the pre-rotation read is 200 and the fresh-auth read is 500 `Failed to decrypt value` ([#18301](https://github.com/OpenHands/OpenHands/issues/18301)). This uses only disposable state and does not establish the original incident's data-loss history.+- **Partial SQLite migration startup (`F26.runtime-services`).** Stop only the owned automation service, make a cold backup of `<run>/private/automation/automations.db`, and confirm its `automation_git_sync_org_config` already has `consecutive_auth_failures`. In that disposable DB only, set `alembic_version.version_num` to `028`; restart with a bounded timeout, then open `/automations`. Expected: required-service startup fails explicitly or the partial schema recovers without corrupting state. Known failure, reproduced 2026-10-12 against automation 1.19.3: migration 029 reports the duplicate column and exits 3, the launcher remains alive, and Canvas shows **Automations Unavailable** ([#18299](https://github.com/OpenHands/OpenHands/issues/18299)). Restore the cold backup, restart and doctor before continuing. Never apply this fixture to an operator database.+- **Occupied internal frontend port (`F26.cli-port-in-use`).** Reserve a loopback frontend port in an owned listener and run the genuine launcher with separate free ingress/backend/automation ports. The reported default-port branch can also be observed read-only when port 3001 is already occupied; do not stop its owner. Expected: an occupied internal frontend port receives a free fallback while explicit service-port conflicts keep their existing semantics. Known failure, reproduced 2026-10-12: `agent-canvas --port 18940` exits 1 with `frontend: port 3001`, despite the ingress override ([#17546](https://github.com/OpenHands/OpenHands/issues/17546), [#16226](https://github.com/OpenHands/OpenHands/issues/16226)). An explicit `OH_CANVAS_SAFE_VITE_PORT` is a current workaround; it does not provide automatic fallback.+- **Required frontend death (`F26.runtime-services`).** After a healthy private launch, identify only its required static-server PID using launcher ancestry plus its owned frontend port, then send that child SIGKILL. Open `/` and check the owned launcher's liveness and authenticated `/api/settings`. Expected: an unexpected required-service death stops the stack and reports failure. Known failure, reproduced 2026-10-12: the launcher and backend remain alive, settings returns 200, and the main UI returns 502 **Bad Gateway** ([#16927](https://github.com/OpenHands/OpenHands/issues/16927)). This live trigger is SIGKILL, not a port-race EADDRINUSE; the failing required-child exit branch is shared. Restart and doctor to restore the run.+- **Static-server credential argv (`F26.runtime-services`).** Inspect only this run's owned static-server process for the frontend port. Read the private key in memory and retain booleans indicating whether `--session-api-key` is present and whether that literal key occurs in argv; never print argv or the key. Expected: other host users cannot recover the session key from command-line arguments. Known failure, reproduced 2026-10-12: both booleans are true for the owned static frontend ([#16383](https://github.com/OpenHands/OpenHands/issues/16383)). Docker remains a separate unavailable prerequisite here.+- **Keep-alive proxy timeout listeners (`F26.runtime-services`).** Through the owned ingress, send 20 sequential authenticated `/api/conversations/search?limit=1` requests on one HTTP keep-alive connection, recording response statuses and the reused local socket port, then inspect only the owned ingress's warning line. Expected: one socket does not acquire one timeout listener per request. Known failure, reproduced 2026-10-12: all 20 requests are 200 on the same socket and ingress emits `MaxListenersExceededWarning` for 11 timeout listeners ([#16839](https://github.com/OpenHands/OpenHands/issues/16839)). `/server_info` uses a custom enrichment handler here, so it is not the generic proxy branch. Keep header values and full logs private.+- **Default Rich logging responsiveness (`F26.remote-backend`).** Temporarily replace only an owned Agent Server with its same pinned interpreter/state on the same port; unset `LOG_JSON`, `CI`, `GITHUB_ACTIONS` and `LOG_RICH_TRACEBACKS`, and retain its PID for cleanup. Create a bare-tool conversation whose dummy LLM points at closed `http://127.0.0.1:9/v1` with zero retries. Start 60 direct `/alive` probes concurrently with a genuine Canvas **Send message**; inspect the rendered connection error and final conversation status. Expected: every probe stays below 1.5 s during failure. Known failure, reproduced 2026-10-12 on Agent Server 1.54.0: the conversation ends `error` and one health probe stalls 2.091 s ([OpenHands/software-agent-sdk#5633](https://github.com/OpenHands/software-agent-sdk/issues/5633)). This is a finite connection-error case, separate from the unprovided cyclic websocket fixture in SDK #5373. Stop that owned replacement, restore the normal launcher and doctor.+## Gotchas- `control-openhands launch` covers only the full stack and `--public`. Partial stacks and `--host` need a hand-started launcher with its own `HOME`, `OH_CANVAS_SAFE_STATE_DIR` and all three `OH_CANVAS_SAFE_*_PORT` variables; without them it writes into the real `~/.openhands/agent-canvas` and grabs the default ports 8000/18000/18001/3001 that another agent may be using.@@ -109,11 +117,11 @@ Preconditions:- The browser daemon refuses other origins unless you pass `--allow-external` to `browser goto`. Each port is its own origin, so a partial-stack page starts at first run again (onboarding and a consent dialog), independent of your run's onboarding.- In a frontend-only UI the **Add a backend** step's Next is enabled with an empty key, while on a LAN or `--public` stack it is disabled until a key is typed: the host the step probes differs (503 vs. auth-required).- A partial-stack origin keeps its stored backend in the browser profile across launches, so a second pass lands straight in the shell. `control-openhands browser reset` brings back the first run on every origin; then `control-openhands browser goto /` and `control-openhands onboard --skip` restore your own run's baseline.-- Which screen a frontend-only origin shows depends on the browser, not the launcher: a profile new to the origin gets **Add a backend** (first run owns the initial backend), while a profile that stores a backend for that origin (the `default-local` entry a full launcher on the same port seeded, or any backend added there earlier) gets the recovery gate of `F25.recovery-gate` as soon as that backend answers 503 (`F26.frontend-only-returning`). The no-toast check belongs to the first-run state only: on the gate, `browser toasts --history` lists generic `An error occurred` status toasts raised by the failing 503 probes (two within 12 s here), with no hint of which request failed. Known failure (reproduced 2026-10-08): both toasts come from the free-models hydrator's `GET /api/llm/models/verified` 503, which no `meta.disableToast` guards on that screen (nor does `backend-version.tsx`); #18160.+- Which screen a frontend-only origin shows depends on the browser, not the launcher: a profile new to the origin gets **Add a backend** (first run owns the initial backend), while a profile that stores a backend for that origin (the `default-local` entry a full launcher on the same port seeded, or any backend added there earlier) gets the recovery gate of `F25.recovery-gate` as soon as that backend answers 503 (`F26.frontend-only-returning`). The no-toast check belongs to the first-run state only: on the gate, `browser toasts --history` lists generic `An error occurred` status toasts raised by the failing 503 probes (two within 12 s here), with no hint of which request failed. Known failure (reproduced 2026-10-08): both toasts come from the free-models hydrator's `GET /api/llm/models/verified` 503, which no `meta.disableToast` guards on that screen (nor does `backend-version.tsx`); [#18160](https://github.com/OpenHands/OpenHands/issues/18160).- Stopping a backend-only instance while the frontend-only page that uses it stays open is the same backend-down state as `F25.recovery-gate` after a reload; keep the returning-user bullet above as the frontend-only-specific check (it reproduces the `default-local` case).- The launcher serves the bundled VS Code editor. #17660 made that opt-in (`OH_CANVAS_ENABLE_VSCODE=true`) and #18048 reverted it, so only a checkout between the two needs the variable (`launch --vscode` sets it). A launcher that serves it reserves `/vscode/`: a frontend-only stack answers `503` there and logs `[static] /vscode -> 503 (rejected)`; a stack with an agent-server lists `/vscode → http://127.0.0.1:<agent-server port + 1000>` in its ingress banner and hands the agent-server that editor port (19961 for the hand-started launchers here, 19000 for the desktop app), which the launcher's own port check does not cover. On a checkout between the two without the variable, `/vscode/` is an ordinary SPA URL (`200`). Here the agent-server has no editor binary (`VSCode server binary not found, VSCode will be disabled` in the launcher log, `/api/vscode/status` `{"running":false,"enabled":true}`), so the editor port stays free even then.-- The port-in-use error is printed twice, the second time with a Node stack trace, and suggests `PORT=<other>` rather than `--port` (#17949). The `--public` key error's example says `npm run dev -- --public` instead of the npx command.-- A backend-only instance's `/server_info.runtime_services` still describes the ingress as routing `/* to the frontend`, which it does not run (#17949).+- The port-in-use error is printed twice, the second time with a Node stack trace, and suggests `PORT=<other>` rather than `--port` ([#17949](https://github.com/OpenHands/OpenHands/issues/17949)). The `--public` key error's example says `npm run dev -- --public` instead of the npx command.+- A backend-only instance's `/server_info.runtime_services` still describes the ingress as routing `/* to the frontend`, which it does not run ([#17949](https://github.com/OpenHands/OpenHands/issues/17949)).- In this sandbox `node_modules/electron/dist` is absent and `node node_modules/electron/install.js` fails with `ERR_ASSERTION` (2026-10-13), so the `F26.desktop-*` rows stay blocked here.- The Electron app hardcodes ports 8000/18000/18001 (and the static server takes 3001), ignores `PORT`, and refuses to start as root without `--no-sandbox` (`Running as root without --no-sandbox is not supported`). `npm run desktop` first runs `build:app`, which rebuilds `build/` under every other run: start `node_modules/.bin/electron --no-sandbox electron` directly against the existing build.- Never stop the desktop app with one group-wide signal: when Xvfb dies with Electron, the app exits without stopping its backend, which is left orphaned (own process groups, parent 1) on 8000/18000/18001/3001 and blocks the next start. If that happens, find the orphans by their ports' command lines (`ps -eo pid,pgid,cmd`, look for `--port 18000`, `--port 18001`, `--port 3001`, `ingress.mjs --port 8000`) and stop each group by id.@@ -121,5 +129,23 @@ Preconditions:- A browser origin keeps the backends it stored on an earlier pass: after the LAN pass on 18960, a later loopback launch on 18960 shows the stale `QA_LAN` backend (red, old key) instead of onboarding, although the page injects a valid key. Assert key-injection bullets through the `window` globals, or `browser reset` first.- Electron's splash shows each raw service-log line (`agent-server: {"asctime": ...`) as its status text during boot; the named phases only appear between them.- `build:lib` from a copy whose `node_modules` is a symlink emits bundled dependencies under `dist/<absolute node_modules path>/...`; that path is an artifact of the copy, not of the product.-- **Bug candidate (`F26.host-bind-lan-optin`):** `scripts/bind-host.mjs` tells npm users to pass `--allow-lan-session-key`, but neither `bin/agent-canvas.mjs` nor `scripts/dev-with-automation.mjs` parses or forwards it (only `scripts/static-server.mjs` and the Docker entrypoint do), and unknown flags are silently ignored (#17949, whose first item is exactly this; reproduced again on 2026-10-08 at d2c89252d).-- Known issue #17946: the npm launcher never creates its `TMUX_TMPDIR`, so hand-started instances share one tmux server and a restart of one resets the others' terminals.+- **Bug candidate (`F26.host-bind-lan-optin`):** `scripts/bind-host.mjs` tells npm users to pass `--allow-lan-session-key`, but neither `bin/agent-canvas.mjs` nor `scripts/dev-with-automation.mjs` parses or forwards it (only `scripts/static-server.mjs` and the Docker entrypoint do), and unknown flags are silently ignored ([#17949](https://github.com/OpenHands/OpenHands/issues/17949), whose first item is exactly this; reproduced again on 2026-10-08 at d2c89252d).+- Known issue [#17946](https://github.com/OpenHands/OpenHands/issues/17946): the npm launcher never creates its `TMUX_TMPDIR`, so hand-started instances share one tmux server and a restart of one resets the others' terminals.+- Known issue [#17763](https://github.com/OpenHands/OpenHands/issues/17763) (`F26.docker-image`): The Docker entrypoint can leave protected Agent Server APIs unauthenticated when only `LOCAL_BACKEND_API_KEY` is supplied, because its export to `OH_SESSION_API_KEYS_0` is inside the generate-key branch. Recheck an owned container with a dummy custom key: unauthenticated `GET /api/conversations/search` must be 401 and the same request with that dummy key 200. `/server_info` is intentionally public and is not an auth probe. Blocked here without a Docker daemon.+- Known issue [#17623](https://github.com/OpenHands/OpenHands/issues/17623) (`F26.helm-chart`): The chart defaults service-account-token automounting to true even with RBAC disabled, exposing an unused cluster credential to agent processes. Recheck a disposable Kubernetes installation using the default values and an explicit opt-in; inspect token existence/readability only, never its contents. Blocked without Helm and a Kubernetes cluster.+- Known issue [#17470](https://github.com/OpenHands/OpenHands/issues/17470) (`F26.docker-image`): On Apple Silicon under Colima/Virtualization.framework, the image's Python services can exit with SIGILL while the frontend leaves onboarding at Disconnected. The report isolates `import litellm` and compares `OPENSSL_armcap=0`; recheck on the affected ARM/Colima guest with the reported cryptography/OpenSSL combination. A native macOS npm stack is not that container environment.+- Known issue [#17112](https://github.com/OpenHands/OpenHands/issues/17112) (`F26.desktop-main-window` `F26.desktop-external-links`): Desktop main-window navigation, an OAuth popup redirect, or a nested popup can render a remote page inside the app frame. Recheck all three on an owned running Electron shell with a harmless remote page and loopback redirect fixture; the browser harness cannot operate Electron windows and the installed Electron binary is unavailable here.+- Known issue [#16872](https://github.com/OpenHands/OpenHands/issues/16872) (`F26.desktop-external-links`): Non-browser URI schemes were reported to reach the OS handler from the desktop popup flow. Current source now parses hosts and restricts external protocols; recheck refusal in a running owned Electron shell before concluding resolution. Observe an inert denied test scheme without invoking any real OS handler. The browser harness cannot drive Electron and no installed Electron binary is available here.+- Known issue [#16774](https://github.com/OpenHands/OpenHands/issues/16774) (`F26.desktop-boot-splash` `F26.desktop-main-window`): The macOS 1.14.0 DMG installs an app that Gatekeeper reports as damaged instead of booting. Recheck the actual downloaded signed/notarized release artifact through install and first launch on macOS; starting source Electron or removing quarantine would bypass the reported failure. That release DMG/signature fixture is unavailable in this run.+- Known issue [#16720](https://github.com/OpenHands/OpenHands/issues/16720) (`F26.runtime-services`): CPU-heavy commands in several conversations can starve Agent Server and stop the UI loading; the report requests lower priority for agent commands than the service. Recheck with a disposable bounded workload, recorded CPU/resource limits and real command observations, never arbitrary stress on the shared machine. The exact Rust workload and affected resource envelope are missing, and real agent execution needs a model key. The backend counterpart is [OpenHands/software-agent-sdk#4575](https://github.com/OpenHands/software-agent-sdk/issues/4575).+- Known issue [#16690](https://github.com/OpenHands/OpenHands/issues/16690) (`F26.docker-image` `F26.runtime-services`): A Docker deployment with nine chats becomes slow/disconnected, then reloads lose the list and startup proxies report ECONNREFUSED. Recheck on the affected Windows-host Docker configuration with retained conversation state and a controlled workload; creating a few empty local chats cannot prove the reported resource/runtime failure. This run lacks Docker and the original conversation/workload fixture.+- Known issue [#16511](https://github.com/OpenHands/OpenHands/issues/16511) (`F26.docker-image`): The documented Docker bind mount can be unwritable to container uid 10001, killing Agent Server/automation while the frontend still serves Bad Gateway. Recheck a disposable host-owned mode-700 mount and a writable control against an owned image; never chmod the operator's real state directory. Blocked here without Docker.+- Known issue [#16300](https://github.com/OpenHands/OpenHands/issues/16300) (`F26.runtime-services`): The npm launcher can fail while building LiteLLM native dependencies when resolved AWS crates require a newer Rust compiler, leaving backend services unavailable. Recheck an isolated uncached uv environment with `litellm==1.95.0`, the reported Rust 1.93.1/toolchain/platform and dependency resolution; a warm current backend cache cannot exercise that install failure. The historical dependency/toolchain fixture is unavailable.+- Known issue [#16208](https://github.com/OpenHands/OpenHands/issues/16208) (`F26.runtime-services`): On Fedora Rawhide/Python 3.14.6, settings work but starting any conversation returns HTTP 500. Recheck a native Fedora environment with the reported SDK/runtime and model/profile, including the underlying server traceback; this macOS current-runtime stack cannot prove that platform-specific failure.+- Known issue [#15434](https://github.com/OpenHands/OpenHands/issues/15434) (`F26.remote-backend`): The embedded VS Code action on a self-hosted reverse-proxy origin was reported to reopen Canvas because its advertised URL omitted the separate editor port/path. Current launchers include an ingress `/vscode/` route; recheck against the affected HTTPS proxy deployment with an actually available editor binary and emitted URL before concluding resolution. This run lacks that self-hosted proxy/editor deployment.+- Known issue [OpenHands/software-agent-sdk#5373](https://github.com/OpenHands/software-agent-sdk/issues/5373) (`F26.remote-backend`): A self-hosted Agent Server using default Rich traceback logging can wedge its event loop and grow memory when a websocket error has a cyclic/deep exception chain; Canvas then times out or disconnects. The npm launcher forces JSON logging, so rechecking needs a separately hosted default-Rich backend plus the reported pathological peer/backlog reproducer and bounded memory/time isolation. That original peer/driver is not included in the issue body; do not synthesize a server patch or stress the shared host and claim a live UI reproduction.+- Known issue [OpenHands/software-agent-sdk#5310](https://github.com/OpenHands/software-agent-sdk/issues/5310) (`F26.docker-conversation-runtime`): A Docker conversation can lose the auxiliary title profile at the runtime boundary, leaving an ACP conversation untitled or producing an unrelated title-auth error while the agent still works. Recheck a saved reachable title profile and an ACP conversation inside an owned per-conversation Docker runtime, comparing its title/profile metadata with the host; inspect only profile names and credential-presence booleans. Blocked without Docker and real auxiliary-title/ACP credentials.+- Known issue [OpenHands/software-agent-sdk#5307](https://github.com/OpenHands/software-agent-sdk/issues/5307) (`F26.docker-conversation-runtime`): New worktree on a Docker-backed conversation can create git metadata whose reverse path exists only inside the container, so the still-live host worktree reads `prunable gitdir file points to non-existent location`. Recheck a disposable real git repository mounted into an owned conversation container, comparing host and container `git worktree list`/status while it is alive; a local worktree does not test the filesystem boundary. Blocked without Docker.+- Known issue [OpenHands/software-agent-sdk#4901](https://github.com/OpenHands/software-agent-sdk/issues/4901) (`F26.runtime-services`): closing an ACP conversation on POSIX is reported to leave the adapter's descendant Node/CLI processes alive after the top npm process exits. Recheck an owned authenticated ACP provider command chain, record its exact descendant identities before close and inspect only those identities afterwards; cleanup must target those owned groups. The current run has no authorized authenticated ACP runtime for this path.+- Known issue [OpenHands/software-agent-sdk#5974](https://github.com/OpenHands/software-agent-sdk/issues/5974) (`F26.runtime-services`; workspace picker in F03): a saved-workspace mutation waiting for its registry file lock can block the Agent Server event loop, including `/alive`, until the lock is released. Recheck only an owned registry using a bounded lock holder, an independently confirmed waiting mutation, concurrent liveness probes and a successful post-release readback; the original evidence uses Linux kernel lock-wait readback. A healthy idle server or concurrent writes without deliberate lock contention do not cover it.+- Known issue [OpenHands/software-agent-sdk#5731](https://github.com/OpenHands/software-agent-sdk/issues/5731) (`F26.cli-public-needs-key`): whitespace-only session keys remain configured credentials in the Agent Server. The Canvas launcher forwards a nonempty `LOCAL_BACKEND_API_KEY` verbatim, so a whitespace-only deployment value can leave the public login unable to use its intended key. Recheck an owned loopback-only public deployment with a whitespace-only supplied key, its actual HTTP/WebSocket authentication and a normal nonblank control; never test the server's default broad bind on an exposed interface. The SDK's missing-`OH_SECRET_KEY` cipher fallback is a separate configuration branch from a Canvas launcher that supplies an encryption key.diff --git a/.agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md b/.agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.mdindex c001b10b4f..75968eeb10 100644--- a/.agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md+++ b/.agents/skills/verify-openhands/references/feature-map/F27-workspace-tools.md@@ -17,7 +17,7 @@ Source: `src/components/features/terminal/`, `src/hooks/use-terminal.ts`, `src/c- `F27.terminal-empty`: with the runtime active and no command run yet, the Terminal tab reads "No terminal output yet. Commands run by the agent will appear here."- `F27.terminal-waiting`: while the runtime is starting or the conversation is in an error state, the Terminal tab reads "Waiting for runtime to start..."; a paused (stopped) conversation shows the empty message instead, by design.- `F27.terminal-output`: the agent's commands appear as `$ <command>` followed by their output; typing into the terminal does nothing (stdin is disabled).-- `F27.terminal-history-reload`: after a page reload the earlier commands and output are still shown (issue #17566).+- `F27.terminal-history-reload`: after a page reload the earlier commands and output are still shown (issue [#17566](https://github.com/OpenHands/OpenHands/issues/17566)).- `F27.terminal-live-append`: a command the agent runs while the page is open is appended live.- `F27.browser-empty`: before any browsing the URL bar reads "No URL loaded", "Open in New Tab" is a disabled button, and the body reads "No page loaded yet. Ask OpenHands to open a URL. Example: "Open https://example.com"".- `F27.browser-screenshot`: after the agent's browser tool ran, the tab shows the last screenshot under a bar with its URL and an enabled "Open in New Tab" link.@@ -26,9 +26,9 @@ Source: `src/components/features/terminal/`, `src/hooks/use-terminal.ts`, `src/c- `F27.planner-plan`: when the workspace has `.agents_tmp/PLAN.md`, the Planner renders it as Markdown with an enabled "Build ⌘↩" bar.- `F27.planner-build`: "Build ⌘↩" sends "Execute the plan based on the .agents_tmp/PLAN.md file." to the agent.- `F27.tasklist`: the Task List tab appears only once the agent used `task_tracker`; it lists each task with its status icon and highlights the in-progress one; the drawer's tabs menu opens it too.-- `F27.usage-metrics`: the Usage tab shows the context meter ("X% used (Y% left)", tokens / window), Token Usage (Input, Cache Hit, Cache Write, Output, Total) and Total Cost.+- `F27.usage-metrics`: the Usage tab shows the context meter ("X% used (Y% left)", tokens / window), Token Usage (Input, Cache Hit, Cache Write, Output, Total) and Total Cost. Above 70% fill, the meter label and bar use the warning feedback role; above 90%, its label uses error feedback and its bar uses status error. Text reaches 4.5:1 against the raised Usage surface and fills reach 3:1 against the track in all five themes.- `F27.usage-empty`: before any metrics arrive the Usage tab reads "No metrics data available".-- `F27.usage-compact`: "Compact context" (agent idle) toasts "Context compaction started" then "Context compacted" (or "…freed N tokens"), and the agent server records a Condensation; the (i) tooltip explains it.+- `F27.usage-compact`: "Compact context" (agent idle) toasts "Context compaction started" then "Context compacted" (or "…freed N tokens"), and the agent server records a Condensation; the (i) tooltip explains it. Above 70% fill, "Context filling up" uses the theme warning feedback color.- `F27.usage-provider-balance`: the provider balance card shows only when the agent server reports a balance (OpenRouter); otherwise it is absent without an error message.- `F27.composer-context-meter`: the composer's context ring opens a popover with the fill label, a Compact context button, tokens and a Usage item; the bar or Usage item opens the Usage tab; Escape (focus returns to the ring) or an outside click closes it.- `F27.overview-peek`: while the drawer is open, hovering the chat header (i) button shows a peek of the Overview rows (with the drawer closed there is no peek; clicking opens the column).@@ -72,7 +72,7 @@ Preconditions:- **Compact context (`F27.usage-compact`).** With the agent idle run `control-openhands browser tooltip 'testid=compact-context-info'` (`Summarizes older messages to free context space while keeping recent details.`), `control-openhands browser mouse-click 1300 700` (closes the tooltip), `control-openhands browser click 'testid=compact-context-button' --observe '[data-testid=compact-context-button]' --observe-ms 4000`, then `control-openhands browser toasts --history` (a plain `browser toasts` right after the click often lists only the first toast: the second lands about 2.5 s later) and `control-openhands conversation events <quiet-id> --last 4`. The last two history toasts are `Context compaction started` and `Context compacted` (a short history frees nothing, so no token count); the events include `CondensationRequest` and `Condensation`.- **Composer context ring (`F27.composer-context-meter`).** Run `control-openhands browser click 'testid=conversation-tab-files'`, `control-openhands browser click 'testid=context-window-meter'` and `control-openhands browser text 'testid=context-window-meter-popover'` (`Context Window`, `1% used (99% left)`, `Compact context`, `11.7k / 1.0M`-style tokens, `Usage`). `control-openhands browser click 'testid=context-window-plan-usage'` then `control-openhands browser count 'testid=usage-panel'` is `1` and the popover count is `0`. Reopen the ring and `control-openhands browser click 'testid=context-window-meter-bar-button'` opens Usage the same way. Reopen it, `control-openhands browser click 'testid=context-window-compact-button'` and, a few seconds later, `control-openhands browser toasts --history` ends with `Context compaction started`, then `Context compacted`; the popover stays open. `control-openhands browser press Escape` closes it (`control-openhands browser count 'testid=context-window-meter-popover'` is `0`) and returns focus to the ring (`control-openhands browser eval "document.activeElement.dataset.testid"` is `context-window-meter`). Reopen it with `control-openhands browser click 'testid=context-window-meter'`; `control-openhands browser mouse-click 600 400` closes it too (`0`).- **Agent-driven tools (`F27.terminal-output`, `F27.canvas-ui-open-tab`).** Run `control-openhands conversation start --prompt "Do these steps in order, in the workspace only. 1) Use the task_tracker tool to set exactly three tasks: 'qa-one' status done, 'qa-two' status in_progress, 'qa-three' status todo. 2) Run the shell command: echo QA_TERM_MARK_1 3) Create the file .agents_tmp/PLAN.md with the content '# QA Plan' followed by a bullet list of three items: alpha, beta, gamma. 4) Call the canvas_ui_control tool with command open_tab and tab terminal. Then reply DONE." --wait --timeout 300` (note `<tools-id>`). The drawer was closed; the agent's tool call opened it: `control-openhands browser attr 'testid=right-panel-toggle' aria-pressed` is `true` and `control-openhands browser text 'testid=tabs-pane-header'` starts with `Terminal`. `control-openhands browser text '.xterm-rows'` shows `$ echo QA_TERM_MARK_1`, `QA_TERM_MARK_1` and the PLAN.md command with its output. `control-openhands browser click '.xterm-screen'`, `control-openhands browser press KeyZ` and `control-openhands browser press Enter` change neither the rows nor the event list (`conversation events <tools-id> --last 1`). `control-openhands conversation events <tools-id> --kinds ActionEvent,ObservationEvent,MessageEvent` shows `open_tab` with `Tool call dispatched to client.`-- **Terminal after reload (`F27.terminal-history-reload`, `F27.terminal-live-append`).** Before clicking any other tab in `<tools-id>` (a user tab click is persisted and would mask the next check), run `control-openhands browser reload`, `control-openhands browser click 'testid=right-panel-toggle'`, `control-openhands browser text 'testid=tabs-pane-header'` (read now for `F27.canvas-ui-open-tab` below; it starts with `Terminal`, so the drawer is already on Terminal: do not click `conversation-tab-terminal`, because clicking the active tab closes the drawer) and `control-openhands browser text '.xterm-rows'`. Expected: the earlier commands are still listed. Known failure (#17566 reproduces, 2026-10-06): the text is empty and the tab reads "No terminal output yet" while the chat still lists the actions. Live append is checked after the Build step below.+- **Terminal after reload (`F27.terminal-history-reload`, `F27.terminal-live-append`).** Before clicking any other tab in `<tools-id>` (a user tab click is persisted and would mask the next check), run `control-openhands browser reload`, `control-openhands browser click 'testid=right-panel-toggle'`, `control-openhands browser text 'testid=tabs-pane-header'` (read now for `F27.canvas-ui-open-tab` below; it starts with `Terminal`, so the drawer is already on Terminal: do not click `conversation-tab-terminal`, because clicking the active tab closes the drawer) and `control-openhands browser text '.xterm-rows'`. Expected: the earlier commands are still listed. Known failure ([#17566](https://github.com/OpenHands/OpenHands/issues/17566) reproduces, 2026-10-06): the text is empty and the tab reads "No terminal output yet" while the chat still lists the actions. Live append is checked after the Build step below.- **Agent tab survives reload (`F27.canvas-ui-open-tab`).** The `tabs-pane-header` text read in the reload step above (right after `right-panel-toggle`) starts with `Terminal`, the agent's last choice: `src/services/canvas-ui.ts` saves the tab per conversation like a user click (#18045), and `control-openhands browser eval "JSON.parse(localStorage.getItem('conversation-state-<tools-id>')).selectedTab"` is `terminal` (`browser screenshot --feature F27.canvas-ui-open-tab --name reload-reopens-terminal`). For contrast, `control-openhands browser click 'testid=conversation-tab-usage'`, `control-openhands browser reload`, `control-openhands browser click 'testid=right-panel-toggle'`: a user-clicked tab is restored (`Usage`).- **Task List (`F27.tasklist`).** In `<tools-id>` run `control-openhands browser click 'testid=conversation-tab-tasklist'`, `control-openhands browser wait '[data-active="true"] >> has-text=qa-two' --timeout 10000` (the rows render a moment after the click; an immediate count reads `0`), `control-openhands browser count '[data-active="true"] >> has-text=qa-two'` (`1`) and `control-openhands browser count '[data-active="false"] >> has-text=qa-three'` (`1`); screenshot with `control-openhands browser screenshot --feature F27.tasklist --name tasks` (done/in-progress/todo icons). Menu entry: `control-openhands browser click 'testid=tabs-pane-header >> testid=ellipsis-button'` then `control-openhands browser click 'testid=conversation-tabs-menu-open-tasklist'`; the header starts with `Task List`.- **Plan and Build (`F27.planner-plan`, `F27.planner-build`).** In `<tools-id>` run `control-openhands browser click 'testid=conversation-tab-planner'` and `control-openhands browser snapshot 'testid=app-route'`: it ends with `heading "QA Plan" [level=1]` and a list `alpha`, `beta`, `gamma`; `control-openhands browser enabled 'testid=planner-tab-build-button'` is `true`. Run `control-openhands browser click 'testid=planner-tab-build-button'`, then `control-openhands browser count 'testid=user-message >> has-text=Execute the plan based on the .agents_tmp/PLAN.md file.'` (`1`) and `control-openhands conversation wait <tools-id> --timeout 240`; the agent reads PLAN.md and answers. Live append (`F27.terminal-live-append`): `control-openhands browser click 'testid=conversation-tab-terminal'` and `control-openhands browser text '.xterm-rows'` now show the new `$ cat .agents_tmp/PLAN.md` block and its output (but not the pre-reload history).@@ -103,8 +103,18 @@ Preconditions:- The Overview peek appears only while the drawer is open (`canPeekOnHover` needs `isRightPanelShown`); a `browser goto` or reload starts with the drawer closed.- The Overview's repo and branch rows depend on a bash probe (`useLocalGitInfo`) that runs only while the conversation's status is idle, running, waiting for confirmation or finished and the agent is not in error. Right after `conversation start` the probe has run and its result stays cached on the page, but on an errored conversation a reload hides the rows again, and `conversation pause` does not apply to an errored conversation. Read the rows before any `goto` or `reload`.- At phone width (and in a narrow drawer) the tab row shows only the tabs that fit; the rest are reachable only through the tabs menu (`conversation-tabs-menu-open-<tab>`), and which tabs fit depends on the active tab's label width.-- Plain `agent-canvas` instances on one machine share the tmux socket `/tmp/tmux-<uid>/openhands` (OpenHands/OpenHands#17946), so another instance stopping can reset this one's terminal. `control-openhands launch` gives each run its own `TMUX_TMPDIR`, so in verification runs a reset means this run's agent server restarted.+- Plain `agent-canvas` instances on one machine share the tmux socket `/tmp/tmux-<uid>/openhands` ([OpenHands/OpenHands#17946](https://github.com/OpenHands/OpenHands/issues/17946)), so another instance stopping can reset this one's terminal. `control-openhands launch` gives each run its own `TMUX_TMPDIR`, so in verification runs a reset means this run's agent server restarted.- Agent Tools & Metadata remembers the last tab (it reopens on Available Tools).- The Overview menu opens only Pull Requests (`conversation-overview-menu-open-git-pull_requests`); its testids are `conversation-overview-menu-open-*` and `-pin-*` for `workspace` and the Git rows only.-- Known open issues: terminal history after reload (#17566) still fails (reproduced 2026-10-06); info modals at phone width (#17562) pass.+- Known open issues: terminal history after reload ([#17566](https://github.com/OpenHands/OpenHands/issues/17566)) still fails (reproduced 2026-10-06); info modals at phone width (#17562) pass.- Agent tab requests (`open_tab`, `navigate_to_file`, chat file links) are saved per conversation like a user tab click (#18045): after the agent opens Terminal or a file and the page reloads, the drawer starts closed and the panel toggle reopens it on that tab; clicking that tab then closes the drawer. At phone width they navigate to `/conversations/<id>/panel`.+- Known issue [#17774](https://github.com/OpenHands/OpenHands/issues/17774) (`F27.planner-empty` `F27.planner-plan`): Planner reads on a native Windows local backend reject drive-letter/mixed-separator workspace paths with `Conversation file path must stay inside the workspace` before a download request. Recheck both a scratch workspace and an opened drive-letter workspace, with and without `.agents_tmp/PLAN.md`; this run lacks a Windows local Agent Server.+- Known issue [#17198](https://github.com/OpenHands/OpenHands/issues/17198) (`F27.terminal-output`): Native Windows desktop agents intermittently return `[Command finished execution with no output]` for real file-listing/terminal commands and retry them. Recheck with the affected Windows installer/runtime, project agent definitions and an exact command/workspace; no deterministic pattern or backend trace is provided, and this run has no Windows desktop runtime. The backend counterpart is [OpenHands/software-agent-sdk#4963](https://github.com/OpenHands/software-agent-sdk/issues/4963).+- Known issue [#16551](https://github.com/OpenHands/OpenHands/issues/16551) (`F27.browser-screenshot`): On native Windows, the browser tool chooses system Chrome over installed Playwright Chromium and times out before returning state/screenshots. Recheck default browser selection and an explicit executable-path control on a Windows host with both browsers and the reported browser_use runtime; the verification Chromium itself is a different tool. Blocked here without Windows and real browser-tool execution credentials.+- Known issue [#16581](https://github.com/OpenHands/OpenHands/issues/16581) (`F27.terminal-output`): A module-scoped terminal command index can suppress output in a second simultaneously mounted Terminal and reset the first when either unmounts. The report describes a future split-panel/embedded concurrent mount rather than a supported standalone entry point. Recheck in a host that actually mounts two independent Canvas Terminal consumers with real command histories; the current harness has one Terminal route and no such host fixture.+- Known issue [#17384](https://github.com/OpenHands/OpenHands/issues/17384) (`F27.launch-child`): A file-based delegated subagent using a saved profile/provider connection can fail with HTTP 401 while the top-level conversation succeeds, because encrypted provider credentials are not decrypted in that backend factory path. Recheck a reviewed `.agents/agents/` worker, encrypted provider connection and saved profile using a real key, comparing parent and worker authentication. This is a delegated SDK subagent path, distinct from the client-tool child-launch recipe; neither a dummy key nor an unencrypted profile proves it.+- Known issue [OpenHands/software-agent-sdk#5533](https://github.com/OpenHands/software-agent-sdk/issues/5533) (`F27.terminal-output`): A slow interactive Bash startup can make the first terminal command return early with empty output and a false successful exit code. Recheck both standalone and pooled tmux using a private `.bashrc` with `sleep 0.4`, then a bounded `sleep 2` first command and its real observation. The shell fixture can be arranged, but the mapped agent-command/Terminal path cannot run here without a model key.+- Known issue [OpenHands/software-agent-sdk#5322](https://github.com/OpenHands/software-agent-sdk/issues/5322) (`F27.terminal-output`): The default tmux pane pool or Windows terminal can let `git log -p` start an interactive pager, so the next agent command receives stale paged output instead of its own result. Recheck a reviewed large git fixture, a real installed pager and a follow-up marker command on each affected terminal backend. The fixture is arrangeable, but the mapped agent-tool path needs a model key (and the Windows half needs Windows).+- Known issue [#17383](https://github.com/OpenHands/OpenHands/issues/17383) (`F27.browser-screenshot`; dependency counterpart [OpenHands/software-agent-sdk#5152](https://github.com/OpenHands/software-agent-sdk/issues/5152)): unlocked dependency resolution can silently remove the agent's browser tools. The actual 2026-10-12 pinned runtime resolves browser-use 0.11.13, mcp 1.30.0 and fastmcp 3.4.8; its genuine BrowserToolExecutor constructs successfully. That narrowly fails to reproduce the SDK report's specified mcp 2.x import failure. Fresh constructor recheck on 2026-10-16 at `b34ac59`, using the actual Agent Server 1.55.0 interpreter and private HOME, again resolves browser-use 0.11.13, mcp 1.30.0 and fastmcp 3.4.8 and constructs successfully; no browser operation or model was run. The Canvas/global browser-operation path still needs an authorized real LLM and a genuine browser-enabled agent issuing a `browser_*` operation; construction alone does not establish that feature works.+- Browser availability also depends on [OpenHands/software-agent-sdk#5899](https://github.com/OpenHands/software-agent-sdk/issues/5899) (`F27.browser-screenshot`; canonical tool-selection note in F13): a Chromium installed only under `PLAYWRIGHT_BROWSERS_PATH` is missed by the agent-tool probe. Recheck the real browser-tool operation on a host with only that custom cache and no standard browser, default cache or PATH executable. This Mac's user Chrome must remain installed; an executor constructor or the verification browser is not a successful agent browser operation.+- Known issue [OpenHands/software-agent-sdk#5835](https://github.com/OpenHands/software-agent-sdk/issues/5835) (`F27.overview-identity`): the local Overview Git probe uses conversation-scoped Bash execution; an idle-eviction pass can kill scoped work without a final output. Recheck a disposable short-idle-TTL server while a genuine Canvas Git probe is still executing, with independently observed command lifetime and the eventual output/result. The report's standalone asynchronous `sleep` reproduction is supporting backend evidence, not proof of a Terminal-tab command, which uses the agent event stream instead.diff --git a/.agents/skills/verify-openhands/references/feature-map/F28-automation-setup-page.md b/.agents/skills/verify-openhands/references/feature-map/F28-automation-setup-page.mdindex 29f21b982d..f61f3734b4 100644--- a/.agents/skills/verify-openhands/references/feature-map/F28-automation-setup-page.md+++ b/.agents/skills/verify-openhands/references/feature-map/F28-automation-setup-page.md@@ -34,7 +34,7 @@ Source: `src/routes/automation-setup-new.tsx`, `src/components/features/automati- `F28.test-gated`: in create mode **Test** stays disabled until the draft is saved and unchanged since; editing any field after a save disables it again until the next **Save draft**.- `F28.draft-test`: **Test** on a saved, valid draft opens **Test runs** (with **Back**), shows `Test run started.` and a Pending row; it runs a hidden disabled copy of the draft, not a dashboard automation. Expected: the row reaches Successful or Failed by itself. Known failure #18263 (reproduced again 2026-10-15 on a Custom Python draft): the row stays Pending and the page does not poll the runs after the first request.- `F28.create`: **Create Automation** creates the automation enabled, navigates to `/automations/<id>` (a Custom Python automation shows **Script**), and removes the draft from the dashboard Drafts group.-- `F28.error-contrast`: the draft bar's error pill and the Test runs error alert use the theme's error color (`text-feedback-error`) on a red tint and reach at least 4.5:1 in all five color themes. The alert does (lowest 4.9:1, Solarized Light); the pill reads about 4.6:1 on Light+ and 4.7:1 on OpenHands-Neutral and OpenHands-Neo. Known failure (reproduced 2026-10-15, [#18481](https://github.com/OpenHands/OpenHands/issues/18481)): with its 10% tint the pill reads 4.33:1 on Solarized Light and 4.48:1 on OpenHands-DeepSea.+- `F28.error-contrast`: the draft bar's error pill and the Test runs error alert use the theme's error color (`text-feedback-error`) and reach at least 4.5:1 in all five color themes. The pill uses a 5% red tint over the draft bar surface.## How to get to it (user POV)@@ -61,7 +61,7 @@ Preconditions:- **Test gating (`F28.test-gated`).** After a successful Save draft `browser enabled 'testid=automation-setup-test'` is `true`; fill `testid=automation-setup-name` with `QA_F28 Probe 2` and it is `false`; Save draft and it is `true`.- **Custom kind and draft Test (`F28.kind-toggle`, `F28.draft-test`).** Click `testid=automation-setup-kind-custom` and `browser wait 'testid=automation-setup-custom-code'`; once the crossfade ends `browser count 'testid=automation-setup-prompt'` is `0` (right after the click it can still read `1`, and `visible` `true`) and `browser value 'testid=automation-setup-custom-code'` contains `AUTOMATION_CALLBACK_URL`. Click `testid=automation-setup-custom-details-toggle`: Entrypoint `python3 main.py`, Setup script path `setup.sh`. Save draft (`Ready to test`), click `testid=automation-setup-test` and `browser wait 'testid=automation-setup-draft-run'`: the row reads Pending and the page shows `Test run started.` Expected: `browser wait 'testid=automation-setup-draft-run >> text=Successful' --timeout 45000`. Known failure #18263 (reproduced 2026-10-15): the row stays Pending for 45 s with a single runs request in `browser network --filter '/runs'`; after `browser reload`, `testid=automation-setup-view-test-runs` shows it Successful. Click `testid=automation-setup-draft-runs-back`. Switching back with `testid=automation-setup-kind-prompt` restores the Prompt; leave the draft on Custom for the next bullet.- **Create (`F28.create`).** Run `control-openhands browser click 'testid=automation-setup-create' --expect-url '/automations/[0-9a-f-]{36}$'`. The detail h1 is `QA_F28 Probe 2`, the badge reads Active and the page shows **Script**. `browser goto /automations`: `browser count 'testid=automation-setup-draft-resume-<draft-id>'` is `0`.-- **Error contrast (`F28.error-contrast`).** Switch to Light+ (`F16`), open a draft whose cron is `every day` (as in the invalid-drafts bullet), `browser screenshot 'testid=automation-setup-draft-details' --feature F28.error-contrast --name light-plus`, and read the colors with `browser eval "getComputedStyle(document.querySelector('[data-testid=automation-setup-draft-validity]')).color"`: `rgb(192, 40, 45)` over `#F3F3F3` with a 10% red tint, about 4.6:1 (at least 4.5:1). Click `testid=automation-setup-test`: the `testid=automation-setup-draft-runs-error` alert has the same color on white, about 5.1:1 (`browser screenshot 'testid=automation-setup-draft-runs-page' --feature F28.error-contrast --name light-plus-runs`). Repeat the pill read on Solarized Light and OpenHands-DeepSea (composite the pill's 10% `#fb2c36` over the bar's `backgroundColor`): expected at least 4.5:1. Known failure (reproduced 2026-10-15, [#18481](https://github.com/OpenHands/OpenHands/issues/18481)): Solarized Light `rgb(187, 42, 38)` over `rgb(238, 232, 213)` is 4.33:1, DeepSea `rgb(231, 106, 94)` over `rgb(33, 37, 47)` is 4.48:1. Restore OpenHands-Neutral.+- **Error contrast (`F28.error-contrast`).** Switch to Light+ (`F16`), open a draft whose cron is `every day` (as in the invalid-drafts bullet), and `browser screenshot 'testid=automation-setup-draft-details' --feature F28.error-contrast --name light-plus`. Read the pill's computed foreground and background and its parent surface with `browser eval`; composite its 5% red tint (`#fb2c36`) over the draft bar background. Expected contrast is at least 4.5:1. Click `testid=automation-setup-test` and inspect `testid=automation-setup-draft-runs-error` the same way. Repeat both checks in OpenHands-DeepSea, OpenHands-Neutral, OpenHands-Neo and Solarized Light, then restore OpenHands-Neutral.- **Cleanup.** Delete `QA_F28 Probe 2` from its kebab and any leftover `QA_F28` draft with its trash icon, as in F21; `api GET /api/automation/v1/drafts` lists none.## Gotchasdiff --git a/.agents/skills/verify-openhands/references/maintenance.md b/.agents/skills/verify-openhands/references/maintenance.mdindex dad9611647..ee9995ee5c 100644--- a/.agents/skills/verify-openhands/references/maintenance.md+++ b/.agents/skills/verify-openhands/references/maintenance.md@@ -165,9 +165,13 @@ automatically a bug and a zero difference does not prove behavior.(reproduced <date>, #<issue>)` note to the bullet that reproduces it, next tothe expected result, or a `Known issue #<issue>` Gotchas entry when no recipecan drive it.-- **Bugs filed outside the pass**: open `bug` issues opened since BASE that- name a mapped feature are repro candidates too. Reproduce each and add its+- **Bugs filed outside the pass**: open `bug` issues without a supported+ map or ledger outcome are repro candidates too, regardless of creation date.+ Include newly opened, reopened and newly bug-labeled reports. Reproduce each and add itsnote; when one does not reproduce, comment on the issue with the run instead.+ `ready-for-dev` prioritizes live work but does not exclude older actionable+ reports. [bug-backfill.md](bug-backfill.md) defines the inventory outcomes and+ their optional GitHub labels; a label must have a map note or report behind it.- **Undocumented or contradictory change**: cite the PR and the observeddifference; ask the maintainer instead of rewriting the expectation.- **Known issues** linked in Gotchas are repro candidates: retest, link freshdiff --git a/.agents/skills/verify-openhands/references/mapping.md b/.agents/skills/verify-openhands/references/mapping.mdindex 0bc6fbd675..3f5a80a49c 100644--- a/.agents/skills/verify-openhands/references/mapping.md+++ b/.agents/skills/verify-openhands/references/mapping.md@@ -130,7 +130,13 @@ sections, in this order:exemptions). The map should know every open bug. Record one on the bulletthat reproduces it, next to the expected result, as `Known failure(reproduced <date>, #<issue>): <what happens>`; list it here as- `Known issue #<issue>: ...` only when no recipe can drive it.+ `Known issue #<issue>: ...` only when no recipe can drive it. The completed+ [existing-bug backfill](backfills/2026-10-12-open-bugs.md) accounts for its frozen+ inventory; the [main refresh](backfills/2026-10-16-main-refresh.md) records later+ reconciliation. [bug-backfill.md](bug-backfill.md) defines ongoing intake and outcome+ labels. Absence of `ready-for-dev` is not an exclusion.+ Use full repository-qualified links for SDK and automation issues, so their+ numbers cannot be mistaken for Canvas issue numbers.Make the note executable when a recipe can drive it: indent a fenced blockunder it, opened with `` ```sh known-bug <ID> [name] `` (the name tellsdiff --git a/.agents/skills/verify-openhands/scripts/browser-daemon.mjs b/.agents/skills/verify-openhands/scripts/browser-daemon.mjsindex 52d01ec779..7bf6d7ab80 100755--- a/.agents/skills/verify-openhands/scripts/browser-daemon.mjs+++ b/.agents/skills/verify-openhands/scripts/browser-daemon.mjs@@ -23,6 +23,10 @@ import { basename, dirname, extname, join, resolve } from "node:path";import { fileURLToPath } from "node:url";import { redactStorage } from "./lib/redact-storage.mjs";import { HeldKeys } from "./lib/held-keys.mjs";+import {+ observeWebSocket,+ isWebSocketObservation,+} from "./lib/websocket-events.mjs";import { SECRET_KEY, SECRET_PATH, redactBody } from "./lib/network-bodies.mjs";import { buildLocator, toCss } from "./lib/selectors.mjs";import {@@ -157,6 +161,7 @@ await context.addInitScript(() => {});let dialogPolicy = "dismiss";+let networkOffline = false;// Events survive a daemon restart (`restart`, `browser reset`): reload the// ones recorded since the last `errors --clear` mark.const events = [];@@ -196,7 +201,10 @@ const heldKeys = new HeldKeys();function originOf(url) {try {- return new URL(url).origin;+ const parsed = new URL(url);+ if (parsed.protocol === "ws:") parsed.protocol = "http:";+ if (parsed.protocol === "wss:") parsed.protocol = "https:";+ return parsed.origin;} catch {return "";}@@ -221,6 +229,12 @@ function record(kind, page, detail) {}function watch(page) {+ page.on("websocket", (socket) =>+ observeWebSocket(socket, (kind, detail) => record(kind, page, detail), {+ connectionId: randomBytes(8).toString("hex"),+ pageUrl: () => page.url(),+ }),+ );page.on("pageerror", (error) =>record("pageerror", page, {message: String(error?.message ?? error).slice(0, 2000),@@ -979,6 +993,7 @@ const handlers = {const relevant = slice.filter((e) =>!["page-opened", "download", "dialog", "mark"].includes(e.kind) &&+ !isWebSocketObservation(e.kind) &&!(noWarnings && e.kind === "console.warning"),);const filtered = appOnly ? relevant.filter((e) => e.appOrigin) : relevant;@@ -1168,6 +1183,18 @@ const handlers = {if (keysOnly !== false) return { area, keys: Object.keys(data).sort() };return { area, storage: redactStorage(data) };},+ async "network-state"({ state }) {+ if (!["offline", "online", "status"].includes(state))+ throw new Error("network-state needs offline | online | status");+ if (state !== "status") {+ const offline = state === "offline";+ if (offline !== networkOffline) {+ await context.setOffline(offline);+ networkOffline = offline;+ }+ }+ return { offline: networkOffline };+ },async network({ clear, external, last, filter, bodies }) {const re = filter ? new RegExp(filter) : null;const list = requests.filter(diff --git a/.agents/skills/verify-openhands/scripts/control-openhands.mjs b/.agents/skills/verify-openhands/scripts/control-openhands.mjsindex ed27a53b0c..e3c40a822f 100755--- a/.agents/skills/verify-openhands/scripts/control-openhands.mjs+++ b/.agents/skills/verify-openhands/scripts/control-openhands.mjs@@ -2821,6 +2821,16 @@ async function cmdBrowser({ positional, flags }) {result = { stopped, profileCleared: true, ...(await startBrowser(run)) };break;}+ case "network-state": {+ const state = rest[0];+ if (rest.length !== 1 || !["offline", "online", "status"].includes(state))+ usage(+ "browser network-state needs offline | online | status",+ "control-openhands browser network-state offline",+ );+ result = await browserCall(run, "network-state", { state });+ break;+ }case "network":result = await browserCall(run, "network", {clear: Boolean(flags.clear),@@ -3890,6 +3900,7 @@ const BROWSER_VERBS = new Set(["tooltip","media","network",+ "network-state","toasts","reset","clipboard",@@ -4755,7 +4766,9 @@ Verbswhose page should see it; the server's clock is untouched)errors [--clear] [--all] [--app-only] [--no-warnings] page/console/HTTP errors since last clear(--clear prints the list, then empties it: run it before the action)- events [--kinds pageerror,dialog,download] [--last N]+ events [--kinds pageerror,dialog,download,websocket-created,websocket-open,websocket-close,websocket-error] [--last N]+ (WebSocket metadata: redacted URLs/pages, numeric after_seq, timestamps and frame counts; no bodies/headers)+ (websocket-open is the first observed frame, not the handshake timestamp; counts update at lifecycle entries)eval <js expression> read-only inspection; never mutate app state with ittabs | tab <i> | tab new [/path] [--allow-external] | close-tab <i> | dialogs [--policy accept|dismiss](tab new: a plain second tab, no opener and no sessionStorage, as a user opening the app again)@@ -4765,6 +4778,8 @@ Verbsstorage [--session] [--values] localStorage (or sessionStorage) keys; values only on request(keys, tokens and secrets inside values are redacted)media [--clear] media playback recorded since load (sound features)+ network-state offline|online|status interrupt or restore this browser context's real network; status is read-only+ (all its tabs; restore online before doctor/cleanup; no request interception)network [--external] [--filter REGEX] [--clear] [--last N] [--bodies] requests with status and redacted query (rows under "recent")(--bodies adds each app-origin write's body, JSON with credential values and env/header mapsreplaced by their length, URL userinfo and secret-named query values redacted in any string;diff --git a/.agents/skills/verify-openhands/scripts/control-openhands.test.mjs b/.agents/skills/verify-openhands/scripts/control-openhands.test.mjsindex 5dd9170713..1cdb6dfde9 100755--- a/.agents/skills/verify-openhands/scripts/control-openhands.test.mjs+++ b/.agents/skills/verify-openhands/scripts/control-openhands.test.mjs@@ -3,15 +3,18 @@// npx vitest run .agents/skills/verify-openhandsimport assert from "node:assert/strict";import { spawn, spawnSync } from "node:child_process";+import { EventEmitter } from "node:events";import {existsSync,mkdirSync,mkdtempSync,readFileSync,readdirSync,+ rmSync,writeFileSync,} from "node:fs";import { createServer } from "node:net";+import { createServer as createHttpServer } from "node:http";import { tmpdir } from "node:os";import { dirname, join, relative, resolve } from "node:path";import { fileURLToPath } from "node:url";@@ -52,6 +55,11 @@ import {} from "./lib/recording.mjs";import { MAX_PAGES, collectEvents, countImages } from "./lib/events-paging.mjs";import { redactBody, redactUrl } from "./lib/network-bodies.mjs";+import {+ observeWebSocket,+ webSocketMetadata,+ isWebSocketObservation,+} from "./lib/websocket-events.mjs";import { buildLocator, parseRole, toCss } from "./lib/selectors.mjs";import {classifyAttempt,@@ -1852,6 +1860,226 @@ test.skipIf(process.platform === "win32")(},);+test("browser network-state restores connectivity and rejects invalid modes before dispatch", async () => {+ // Arrange a browser service. The CLI must route state changes without+ // confusing the observation-only network log with network availability.+ const runDir = mkdtempSync(join(tmpdir(), "cov-network-state-"));+ mkdirSync(join(runDir, "private"));+ writeFileSync(join(runDir, "run.json"), "{}\n");+ let offline = false;+ let requests = 0;+ const server = createHttpServer(async (req, res) => {+ requests += 1;+ let body = "";+ for await (const chunk of req) body += chunk;+ const { cmd, args } = JSON.parse(body);+ assert.equal(cmd, "network-state");+ if (args.state !== "status") offline = args.state === "offline";+ res.setHeader("content-type", "application/json");+ res.end(JSON.stringify({ ok: true, offline }));+ });+ await new Promise((resolve) => server.listen(0, "127.0.0.1", resolve));+ writeFileSync(+ join(runDir, "private", "browser.json"),+ JSON.stringify({+ pid: process.pid,+ port: server.address().port,+ token: "dummy-control-token",+ }),+ );+ const invoke = (state) =>+ new Promise((resolve, reject) => {+ const child = spawn(process.execPath, [+ cli,+ "browser",+ "network-state",+ ...state,+ "--run",+ runDir,+ ]);+ let output = "";+ child.stdout.on("data", (chunk) => {+ output += chunk;+ });+ child.on("error", reject);+ child.on("close", (code) => resolve({ code, ...JSON.parse(output) }));+ });+ try {+ // Act / Assert: status is read-only, and online restores the prior+ // offline browser. Bad input must not reach or mutate the service.+ assert.equal((await invoke(["offline"])).offline, true);+ assert.equal((await invoke(["status"])).offline, true);+ assert.equal((await invoke(["online"])).offline, false);+ const before = requests;+ for (const args of [[], ["sometimes"], ["offline", "extra"]]) {+ assert.equal((await invoke(args)).code, 2);+ }+ assert.equal(requests, before);+ } finally {+ await new Promise((resolve) => server.close(resolve));+ rmSync(runDir, { recursive: true, force: true });+ }+});++test("WebSocket subscription events preserve cursors and counts without secrets or frame content", () => {+ // Arrange the underlying Playwright socket event source; the real observer+ // must expose only metadata to browser-events.jsonl and browser events.+ const socket = new EventEmitter();+ socket.url = () =>+ "ws://user:SOCKET_PASSWORD@localhost:18800/sockets/events/qa?after_seq=7&session_api_key=SESSION_SECRET&unknown=OPAQUE_SECRET";+ const events = [];+ observeWebSocket(socket, (kind, detail) => events.push({ kind, ...detail }), {+ connectionId: "qa-connection",+ pageUrl: () =>+ "http://user:PAGE_PASSWORD@localhost:18800/conversations/qa?api_key=PAGE_SECRET#TOKEN_SECRET",+ });+ // Act: even authentication frames and error text must never be retained.+ socket.emit("framereceived", {+ payload: "RECEIVED_SECRET",+ headers: { authorization: "HEADER_SECRET" },+ });+ socket.emit("framesent", { payload: Buffer.from("SENT_SECRET") });+ socket.emit("socketerror", "ERROR_SECRET");+ socket.emit("close");+ // Assert the consumer contract, including healthy/error classification.+ assert.equal(events[0].afterSeq, "7");+ assert.equal(new URL(events[0].url).searchParams.get("after_seq"), "7");+ const opened = events.find((e) => e.kind === "websocket-open");+ assert.equal(opened.observed, "first-frame");+ assert.equal(opened.receivedFrames, 1);+ const closed = events.at(-1);+ assert.equal(closed.receivedFrames, 1);+ assert.equal(closed.sentFrames, 1);+ assert.equal(closed.connectionId, events[0].connectionId);+ assert.deepEqual(+ events.filter((e) => !isWebSocketObservation(e.kind)).map((e) => e.kind),+ ["websocket-error"],+ );+ const shown = JSON.stringify(events);+ for (const secret of [+ "SOCKET_PASSWORD",+ "SESSION_SECRET",+ "OPAQUE_SECRET",+ "PAGE_PASSWORD",+ "PAGE_SECRET",+ "TOKEN_SECRET",+ "RECEIVED_SECRET",+ "HEADER_SECRET",+ "SENT_SECRET",+ "ERROR_SECRET",+ ])+ assert.ok(!shown.includes(secret));+ assert.equal(+ webSocketMetadata("ws://localhost/x?after_seq=NOT_A_CURSOR_SECRET")+ .afterSeq,+ undefined,+ );+ assert.ok(+ !JSON.stringify(+ webSocketMetadata("ws://localhost/x?after_seq=NOT_A_CURSOR_SECRET"),+ ).includes("NOT_A_CURSOR_SECRET"),+ );+ assert.ok(+ !JSON.stringify(+ webSocketMetadata(+ "ws://localhost/x?after_seq=7&after_seq=DUPLICATE_SECRET",+ ),+ ).includes("DUPLICATE_SECRET"),+ );+ assert.equal(+ webSocketMetadata("ws://localhost/x?after_seq=7&after_seq=8").afterSeq,+ undefined,+ );+ assert.ok(+ !JSON.stringify(+ webSocketMetadata("ws://localhost/x?SECRET_NAME=ignored"),+ ).includes("SECRET_NAME"),+ );+ assert.deepEqual(webSocketMetadata("invalid SECRET"), {+ url: "<unavailable>",+ });+});++test("browser events delivers filtered WebSocket metadata to the CLI consumer", async () => {+ const runDir = mkdtempSync(join(tmpdir(), "cov-websocket-events-"));+ mkdirSync(join(runDir, "private"));+ writeFileSync(join(runDir, "run.json"), "{}\n");+ const socket = new EventEmitter();+ socket.url = () =>+ "ws://localhost:18800/sockets/events/qa?after_seq=42&session_api_key=PRIVATE_SECRET";+ const events = [];+ observeWebSocket(+ socket,+ (kind, detail) =>+ events.push({ ts: new Date().toISOString(), kind, ...detail }),+ {+ connectionId: "qa-reconnect",+ pageUrl: () => "http://localhost:18800/conversations/qa",+ },+ );+ socket.emit("framesent", { payload: "AUTH_SECRET" });+ socket.emit("close");+ const server = createHttpServer(async (req, res) => {+ let body = "";+ for await (const chunk of req) body += chunk;+ const { cmd, args } = JSON.parse(body);+ assert.equal(cmd, "events");+ assert.equal(args.kinds, "websocket-open,websocket-close");+ assert.equal(args.last, "2");+ const wanted = args.kinds.split(",");+ res.setHeader("content-type", "application/json");+ res.end(+ JSON.stringify({+ ok: true,+ events: events+ .filter((e) => wanted.includes(e.kind))+ .slice(-Number(args.last)),+ }),+ );+ });+ await new Promise((resolve) => server.listen(0, "127.0.0.1", resolve));+ writeFileSync(+ join(runDir, "private", "browser.json"),+ JSON.stringify({+ pid: process.pid,+ port: server.address().port,+ token: "dummy-control-token",+ }),+ );+ try {+ const output = await new Promise((resolve, reject) => {+ const child = spawn(process.execPath, [+ cli,+ "browser",+ "events",+ "--kinds",+ "websocket-open,websocket-close",+ "--last",+ "2",+ "--run",+ runDir,+ ]);+ let stdout = "";+ child.stdout.on("data", (chunk) => {+ stdout += chunk;+ });+ child.on("error", reject);+ child.on("close", (code) => resolve({ code, stdout }));+ });+ assert.equal(output.code, 0);+ const result = JSON.parse(output.stdout);+ assert.equal(result.events.length, 2);+ assert.equal(result.events[0].afterSeq, "42");+ assert.equal(result.events[1].sentFrames, 1);+ assert.ok(result.events.every((e) => Number.isFinite(Date.parse(e.ts))));+ assert.ok(!output.stdout.includes("PRIVATE_SECRET"));+ assert.ok(!output.stdout.includes("AUTH_SECRET"));+ } finally {+ await new Promise((resolve) => server.close(resolve));+ rmSync(runDir, { recursive: true, force: true });+ }+});+// A run directory with no stack behind it: enough for evidence and bug runs// whose recipes only call offline commands.function fakeRun(revision = "abc") {diff --git a/.agents/skills/verify-openhands/scripts/fixtures/accept-only-endpoint.py b/.agents/skills/verify-openhands/scripts/fixtures/accept-only-endpoint.pynew file mode 100755index 0000000000..f3594b4a21--- /dev/null+++ b/.agents/skills/verify-openhands/scripts/fixtures/accept-only-endpoint.py@@ -0,0 +1,61 @@+#!/usr/bin/env python3+"""Bounded loopback TCP endpoint that accepts connections and never answers.++No request bytes, headers, or credentials are read or recorded. Use a dummy+model/key in the caller. The private port.json records only port/PID/deadline.+"""++import argparse+import json+import os+from pathlib import Path+import signal+import socket+import time+++def arguments():+ p = argparse.ArgumentParser(description=__doc__)+ p.add_argument("--output-dir", type=Path, required=True)+ p.add_argument("--seconds", type=int, default=180)+ a = p.parse_args()+ if not 1 <= a.seconds <= 180:+ p.error("seconds must be within1..180")+ return a+++if __name__ == "__main__":+ a = arguments()+ a.output_dir.mkdir(parents=True, exist_ok=True)+ os.chmod(a.output_dir, 0o700)+ held = []+ began = time.time()+ server = socket.create_server(("127.0.0.1", 0))+ server.settimeout(0.5)+ port = server.getsockname()[1]+ metadata = {"pid": os.getpid(), "port": port, "started_at": began,+ "deadline": began + a.seconds, "received_bytes_recorded": 0}+ target = a.output_dir / "port.json"+ target.write_text(json.dumps(metadata, indent=2))+ os.chmod(target, 0o600)+ def stop(_sig, _frame):+ raise SystemExit(0)+ signal.signal(signal.SIGTERM, stop)+ signal.signal(signal.SIGINT, stop)+ try:+ while time.time() < metadata["deadline"] and len(held) < 32:+ try:+ connection, _address = server.accept()+ held.append(connection)+ metadata["accepted_connections"] = len(held)+ target.write_text(json.dumps(metadata, indent=2))+ except TimeoutError:+ pass+ finally:+ for connection in held:+ connection.close()+ server.close()+ metadata["closed_at"] = time.time()+ metadata["accepted_connections"] = len(held)+ target.write_text(json.dumps(metadata, indent=2))+ print(json.dumps({"port": port, "accepted_connections": len(held), "closed": True}))diff --git a/.agents/skills/verify-openhands/scripts/fixtures/download-trajectory-overlap.py b/.agents/skills/verify-openhands/scripts/fixtures/download-trajectory-overlap.pynew file mode 100755index 0000000000..a8888cf7cb--- /dev/null+++ b/.agents/skills/verify-openhands/scripts/fixtures/download-trajectory-overlap.py@@ -0,0 +1,169 @@+#!/usr/bin/env python3+"""Bounded real-loopback trajectory downloads; only sanitized counts are printed.++Run with the SDK environment's Python (which includes httpx). Raw ZIP samples+stay in a private output directory. No response is mocked or replayed.+"""++import argparse+import asyncio+from collections import Counter+import io+import ipaddress+import json+import os+from pathlib import Path+import time+from urllib.parse import urlsplit+from uuid import UUID+import zipfile+import zlib+++def arguments(argv=None):+ p = argparse.ArgumentParser(description=__doc__)+ p.add_argument("--base-url", required=True)+ p.add_argument("--conversation-id", required=True)+ p.add_argument("--api-key-file", type=Path, required=True)+ p.add_argument("--output-dir", type=Path, required=True)+ p.add_argument("--seconds", type=float, default=30)+ p.add_argument("--clients", type=int, default=3)+ p.add_argument("--rounds", type=int, default=2000, help="Maximum requests per client")+ p.add_argument("--slow-ms", type=float, default=0, help="Delay after each received chunk")+ p.add_argument("--max-bytes", type=int, default=16 * 1024 * 1024)+ p.add_argument("--max-uncompressed-bytes", type=int, default=16 * 1024 * 1024)+ p.add_argument("--max-members", type=int, default=512)+ a = p.parse_args(argv)+ u = urlsplit(a.base_url)+ try:+ local = ipaddress.ip_address(u.hostname or "").is_loopback+ port = u.port+ except ValueError:+ local, port = False, None+ if (u.scheme != "http" or not local or not port or u.username or u.password+ or u.query or u.fragment or u.path not in ("", "/")):+ p.error("base-url must be a literal loopback HTTP origin with a port")+ try:+ a.conversation_id = str(UUID(a.conversation_id))+ except ValueError:+ p.error("conversation-id must be a UUID")+ if not (1 <= a.seconds <= 60 and 1 <= a.clients <= 4+ and 1 <= a.rounds <= 5000 and 0 <= a.slow_ms <= 250+ and 1024 <= a.max_bytes <= 16 * 1024 * 1024+ and 1024 <= a.max_uncompressed_bytes <= 16 * 1024 * 1024+ and 1 <= a.max_members <= 512):+ p.error("bounds: seconds1..60, clients1..4, rounds1..5000, slow-ms0..250, max-bytes1KiB..16MiB")+ return a+++def zip_outcome(body, *, max_uncompressed_bytes=16 * 1024 * 1024,+ max_members=512, deadline=float("inf")):+ try:+ if time.monotonic() >= deadline:+ return "unverified_deadline"+ with zipfile.ZipFile(io.BytesIO(body)) as archive:+ members = archive.infolist()+ if len(members) > max_members:+ return "unverified_member_limit"+ if sum(info.file_size for info in members) > max_uncompressed_bytes:+ return "unverified_uncompressed_limit"+ total = 0+ for info in members:+ if time.monotonic() >= deadline:+ return "unverified_deadline"+ if info.is_dir():+ continue+ with archive.open(info) as member:+ while True:+ if time.monotonic() >= deadline:+ return "unverified_deadline"+ chunk = member.read(min(32768, max_uncompressed_bytes - total + 1))+ total += len(chunk)+ if total > max_uncompressed_bytes:+ return "unverified_uncompressed_limit"+ if not chunk:+ break+ return "valid"+ except (NotImplementedError, RuntimeError):+ return "unverified_unsupported_archive"+ except (zipfile.BadZipFile, OSError, EOFError, zlib.error):+ return "broken"+++async def measure(a):+ import httpx++ key = a.api_key_file.read_text().strip()+ if not key:+ raise ValueError("The session-key file is empty")+ a.output_dir.mkdir(parents=True, exist_ok=True)+ os.chmod(a.output_dir, 0o700)+ url = a.base_url.rstrip("/") + "/api/file/download-trajectory/" + a.conversation_id+ started = time.monotonic()+ deadline = started + a.seconds++ async def receiver(index):+ counts = Counter()+ samples = set()+ async with httpx.AsyncClient(+ headers={"X-Session-API-Key": key}, trust_env=False,+ follow_redirects=False, timeout=5,+ limits=httpx.Limits(max_connections=1, max_keepalive_connections=1),+ ) as client:+ for _ in range(a.rounds):+ if time.monotonic() >= deadline:+ break+ counts["attempts"] += 1+ body = bytearray()+ try:+ async with client.stream("GET", url) as response:+ counts["http_" + str(response.status_code)] += 1+ abort_reason = None+ async for chunk in response.aiter_bytes(chunk_size=32768):+ if time.monotonic() >= deadline:+ abort_reason = "unverified_deadline"+ break+ body.extend(chunk)+ if len(body) > a.max_bytes:+ abort_reason = "unverified_network_byte_limit"+ break+ if a.slow_ms:+ await asyncio.sleep(a.slow_ms / 1000)+ if response.status_code != 200:+ continue+ length = response.headers.get("content-length")+ if abort_reason is None and length and int(length) != len(body):+ counts["length_mismatch"] += 1+ label = abort_reason or zip_outcome(+ body, max_uncompressed_bytes=a.max_uncompressed_bytes,+ max_members=a.max_members, deadline=deadline)+ counts[label + "_zip" if label in ("valid", "broken") else label] += 1+ counts["received_bytes"] += len(body)+ if label not in samples:+ raw = a.output_dir / f"client-{index}-{label}.zip"+ raw.write_bytes(body)+ os.chmod(raw, 0o600)+ samples.add(label)+ except httpx.HTTPError as error:+ counts["transport_" + type(error).__name__] += 1+ return counts++ combined = Counter()+ for counts in await asyncio.gather(*(receiver(i) for i in range(a.clients))):+ combined.update(counts)+ summary = {+ "origin": a.base_url.rstrip("/"), "clients": a.clients,+ "duration_limit_seconds": a.seconds, "round_limit_per_client": a.rounds,+ "slow_ms": a.slow_ms, "elapsed_seconds": round(time.monotonic() - started, 3),+ "max_network_bytes": a.max_bytes, "max_uncompressed_bytes": a.max_uncompressed_bytes,+ "max_members": a.max_members,+ "counts": dict(combined), "raw_samples_private": True,+ }+ target = a.output_dir / "summary.json"+ target.write_text(json.dumps(summary, indent=2) + "\n")+ os.chmod(target, 0o600)+ return summary+++if __name__ == "__main__":+ print(json.dumps(asyncio.run(measure(arguments())), indent=2))diff --git a/.agents/skills/verify-openhands/scripts/fixtures/shared-storage-server.py b/.agents/skills/verify-openhands/scripts/fixtures/shared-storage-server.pynew file mode 100755index 0000000000..0b076e1c0c--- /dev/null+++ b/.agents/skills/verify-openhands/scripts/fixtures/shared-storage-server.py@@ -0,0 +1,125 @@+#!/usr/bin/env python3+"""Bounded no-editor Agent Server fixture sharing one verification run's storage.++Use the installed SDK's Python. Keys are read only from the run's private files;+they remain in the child's environment and are never printed. Stop this parent+with SIGTERM for deterministic child-process-group cleanup.+"""++import argparse+import json+import os+from pathlib import Path+import signal+import socket+import subprocess+import sys+import time+from uuid import UUID+++def group_alive(pgid):+ try:+ os.killpg(pgid, 0)+ return True+ except ProcessLookupError:+ return False+++def stop_owned_group(child, grace_seconds=5, force_seconds=2):+ pgid = child.pid # Popen(start_new_session=True) established this owned group.+ forced = False+ try:+ os.killpg(pgid, signal.SIGTERM)+ except ProcessLookupError:+ pass+ deadline = time.monotonic() + grace_seconds+ while group_alive(pgid) and time.monotonic() < deadline:+ child.poll() # Reap the leader without mistaking it for the whole group.+ time.sleep(0.05)+ if group_alive(pgid):+ forced = True+ try:+ os.killpg(pgid, signal.SIGKILL)+ except ProcessLookupError:+ pass+ deadline = time.monotonic() + force_seconds+ while group_alive(pgid) and time.monotonic() < deadline:+ child.poll()+ time.sleep(0.05)+ child.poll()+ return {"leader_stopped": child.returncode is not None,+ "group_stopped": not group_alive(pgid), "forced": forced}+++if __name__ == "__main__":+ p = argparse.ArgumentParser(description=__doc__)+ p.add_argument("--run", type=Path, required=True)+ p.add_argument("--seconds", type=int, default=240)+ p.add_argument("--load-id", type=UUID, help="Read native events/count on the existing owned secondary")+ a = p.parse_args()+ if not 1 <= a.seconds <= 300:+ p.error("seconds must be within1..300")+ run = a.run.resolve()+ json.loads((run / "run.json").read_text())+ directory = run / "private/shared-server-5629"+ directory.mkdir(exist_ok=True)+ os.chmod(directory, 0o700)+ if a.load_id:+ import httpx+ metadata = json.loads((directory / "server.json").read_text())+ if metadata.get("stopped_at") or time.time() >= metadata["deadline"]:+ p.error("the bounded owned secondary is stopped or expired")+ os.kill(metadata["pid"], 0)+ port = int(metadata["port"])+ if not 1024 <= port <= 65535:+ p.error("invalid owned secondary port")+ with httpx.Client(trust_env=False, follow_redirects=False, timeout=5) as client:+ response = client.get(+ f"http://127.0.0.1:{port}/api/conversations/{a.load_id}/events/count",+ headers={"X-Session-API-Key": (run / "private/session-key").read_text().strip()},+ )+ print(json.dumps({"status": response.status_code,+ "event_count": response.json() if response.status_code == 200 else None}))+ raise SystemExit(0)+ env = json.loads((run / "private/launcher-env.json").read_text())+ env.update({+ "OH_SECRET_KEY": (run / "private/encryption-key").read_text().strip(),+ "OH_SESSION_API_KEYS_0": (run / "private/session-key").read_text().strip(),+ "OH_CONVERSATIONS_PATH": str(run / "private/state/dev_conversations"),+ "OH_WORKSPACE_PATH": str(run / "workspace/qa-rechecks"),+ "OH_BASH_EVENTS_DIR": str(directory / "bash-events"),+ "OH_ENABLE_VSCODE": "false",+ "OH_PERSISTENCE_DIR": str(directory / "state"),+ })+ with socket.create_server(("127.0.0.1", 0)) as probe:+ port = probe.getsockname()[1]+ stopped = False+ def stop(_sig, _frame):+ global stopped+ stopped = True+ signal.signal(signal.SIGTERM, stop)+ signal.signal(signal.SIGINT, stop)+ with (directory / "server.log").open("wb") as log:+ metadata = {}+ target = directory / "server.json"+ child = subprocess.Popen(+ [sys.executable, "-m", "openhands.agent_server", "--host", "127.0.0.1",+ "--port", str(port)], env=env, cwd=directory,+ stdout=log, stderr=subprocess.STDOUT, start_new_session=True,+ )+ try:+ metadata = {"pid": child.pid, "pgid": child.pid, "operator_pid": os.getpid(),+ "port": port, "deadline": time.time() + a.seconds,+ "shared_storage": str(run / "private/state/dev_conversations")}+ target.write_text(json.dumps(metadata, indent=2))+ os.chmod(target, 0o600)+ while not stopped and child.poll() is None and time.time() < metadata["deadline"]:+ time.sleep(0.2)+ finally:+ cleanup = stop_owned_group(child)+ metadata.update({"stopped_at": time.time(), "exit_code": child.returncode,+ "cleanup": cleanup})+ target.write_text(json.dumps(metadata, indent=2))+ print(json.dumps({"port": port, **cleanup,+ "exit_code": child.returncode}))diff --git a/.agents/skills/verify-openhands/scripts/fixtures/test_download_trajectory_overlap.py b/.agents/skills/verify-openhands/scripts/fixtures/test_download_trajectory_overlap.pynew file mode 100644index 0000000000..8b72567e87--- /dev/null+++ b/.agents/skills/verify-openhands/scripts/fixtures/test_download_trajectory_overlap.py@@ -0,0 +1,125 @@+"""Safety contract for the real local trajectory receiver fixture."""++from contextlib import redirect_stderr+import importlib.util+import io+from pathlib import Path+import unittest+import zipfile+import struct+import time+import asyncio+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer+import tempfile+import threading+++spec = importlib.util.spec_from_file_location(+ "receiver", Path(__file__).with_name("download-trajectory-overlap.py")+)+receiver = importlib.util.module_from_spec(spec)+spec.loader.exec_module(receiver)+++class ReceiverSafetyTest(unittest.TestCase):+ def args(self, origin="http://127.0.0.1:18810", extras=()):+ return ["--base-url", origin, "--conversation-id",+ "00000000-0000-0000-0000-000000000001", "--api-key-file",+ "/unused/private/session-key", "--output-dir", "/unused/private/output", *extras]++ def test_rejects_nonloopback_and_credential_bearing_origins_before_io(self):+ for origin in ("http://example.com:80", "http://localhost:80",+ "http://127.0.0.1:[email protected]:80", "https://127.0.0.1:80",+ "http://127.0.0.1:80/?key=private", "http://127.0.0.1:80/#private"):+ with self.subTest(origin=origin), redirect_stderr(io.StringIO()):+ with self.assertRaises(SystemExit) as error:+ receiver.arguments(self.args(origin))+ self.assertEqual(error.exception.code, 2)+ self.assertEqual(receiver.arguments(self.args()).clients, 3)+ self.assertEqual(receiver.arguments(self.args("http://[::1]:80")).base_url,+ "http://[::1]:80")++ def test_rejects_unbounded_duration_concurrency_and_body_sizes(self):+ for extra in (("--seconds", "61"), ("--clients", "5"),+ ("--rounds", "5001"), ("--slow-ms", "251"),+ ("--max-bytes", str(17 * 1024 * 1024))):+ with self.subTest(extra=extra), redirect_stderr(io.StringIO()):+ with self.assertRaises(SystemExit) as error:+ receiver.arguments(self.args(extras=extra))+ self.assertEqual(error.exception.code, 2)++ def test_a_broken_compressed_member_is_counted_without_aborting_receivers(self):+ buf = io.BytesIO()+ with zipfile.ZipFile(buf, "w", compression=zipfile.ZIP_DEFLATED) as archive:+ archive.writestr("fixture.txt", "bounded fixture content" * 100)+ original = buf.getvalue()+ self.assertEqual(receiver.zip_outcome(original), "valid")+ damaged = bytearray(original)+ name_length, extra_length = struct.unpack_from("<HH", damaged, 26)+ start = 30 + name_length + extra_length+ damaged[start:start + 8] = b"\xff" * 8+ self.assertEqual(receiver.zip_outcome(damaged), "broken")++ def test_validation_budgets_and_deadlines_are_unverified(self):+ buf = io.BytesIO()+ with zipfile.ZipFile(buf, "w", compression=zipfile.ZIP_DEFLATED) as archive:+ archive.writestr("fixture.txt", "x" * 100000)+ archive.writestr("second.txt", "small")+ body = buf.getvalue()+ self.assertEqual(receiver.zip_outcome(body, max_uncompressed_bytes=1024),+ "unverified_uncompressed_limit")+ self.assertEqual(receiver.zip_outcome(body, max_members=1), "unverified_member_limit")+ self.assertEqual(receiver.zip_outcome(body, deadline=time.monotonic() - 1),+ "unverified_deadline")++ def test_unsupported_compression_is_unverified(self):+ buf = io.BytesIO()+ with zipfile.ZipFile(buf, "w") as archive:+ archive.writestr("fixture.txt", "small")+ body = bytearray(buf.getvalue())+ struct.pack_into("<H", body, 8, 99)+ central = body.index(b"PK\x01\x02")+ struct.pack_into("<H", body, central + 10, 99)+ self.assertEqual(receiver.zip_outcome(body), "unverified_unsupported_archive")++ def test_intentional_network_cap_does_not_claim_transport_truncation(self):+ archive = io.BytesIO()+ with zipfile.ZipFile(archive, "w") as z:+ z.writestr("bounded.txt", "x" * 65536)+ body = archive.getvalue()+ class Handler(BaseHTTPRequestHandler):+ def do_GET(self):+ self.send_response(200)+ self.send_header("Content-Length", str(len(body)))+ self.end_headers()+ try:+ self.wfile.write(body)+ except (BrokenPipeError, ConnectionResetError):+ pass+ def log_message(self, *_args):+ pass+ service = ThreadingHTTPServer(("127.0.0.1", 0), Handler)+ thread = threading.Thread(target=service.serve_forever, daemon=True)+ thread.start()+ try:+ with tempfile.TemporaryDirectory() as directory:+ key = Path(directory) / "key"+ key.write_text("dummy-test-only")+ args = receiver.arguments([+ "--base-url", f"http://127.0.0.1:{service.server_port}",+ "--conversation-id", "00000000-0000-0000-0000-000000000001",+ "--api-key-file", str(key), "--output-dir", str(Path(directory) / "samples"),+ "--seconds", "1", "--clients", "1", "--rounds", "1", "--max-bytes", "1024"])+ summary = asyncio.run(receiver.measure(args))+ self.assertEqual(summary["counts"]["unverified_network_byte_limit"], 1)+ self.assertNotIn("length_mismatch", summary["counts"])+ self.assertNotIn("valid_zip", summary["counts"])+ self.assertNotIn("broken_zip", summary["counts"])+ finally:+ service.shutdown()+ service.server_close()+ thread.join(timeout=1)+++if __name__ == "__main__":+ unittest.main()diff --git a/.agents/skills/verify-openhands/scripts/fixtures/test_shared_storage_server.py b/.agents/skills/verify-openhands/scripts/fixtures/test_shared_storage_server.pynew file mode 100644index 0000000000..bb82e369e7--- /dev/null+++ b/.agents/skills/verify-openhands/scripts/fixtures/test_shared_storage_server.py@@ -0,0 +1,59 @@+"""Real child-group cleanup contract for the bounded SDK fixture."""++import importlib.util+import os+from pathlib import Path+import subprocess+import signal+import sys+import tempfile+import time+import unittest++spec = importlib.util.spec_from_file_location(+ "shared_server", Path(__file__).with_name("shared-storage-server.py")+)+server = importlib.util.module_from_spec(spec)+spec.loader.exec_module(server)++[email protected](os.name == "posix", "The fixture requires POSIX process groups")+class GroupCleanupTest(unittest.TestCase):+ def check_orphan(self, ignore_term):+ with tempfile.TemporaryDirectory() as directory:+ ready = Path(directory) / "ready"+ worker = ("import signal,time,pathlib,sys; "+ + ("signal.signal(signal.SIGTERM,signal.SIG_IGN); " if ignore_term else "")+ + "pathlib.Path(sys.argv[1]).touch(); time.sleep(20)")+ parent = "import subprocess,sys; subprocess.Popen([sys.executable,'-c',sys.argv[1],sys.argv[2]])"+ child = subprocess.Popen([sys.executable, "-c", parent, worker, str(ready)],+ start_new_session=True)+ try:+ child.wait(timeout=3)+ deadline = time.monotonic() + 3+ while not ready.exists() and time.monotonic() < deadline:+ time.sleep(0.02)+ self.assertTrue(ready.exists())+ self.assertTrue(server.group_alive(child.pid))+ result = server.stop_owned_group(child, grace_seconds=.15, force_seconds=2)+ self.assertTrue(result["leader_stopped"])+ self.assertTrue(result["group_stopped"])+ self.assertEqual(result["forced"], ignore_term)+ finally:+ server.stop_owned_group(child, grace_seconds=.05, force_seconds=2)+ # Independent test cleanup also protects the deliberately failing+ # pre-fix regression run; assertions above test the helper itself.+ try:+ os.killpg(child.pid, signal.SIGKILL)+ except ProcessLookupError:+ pass++ def test_descendant_is_stopped_after_leader_already_exited(self):+ self.check_orphan(False)++ def test_ignoring_descendant_is_forced_after_bounded_grace(self):+ self.check_orphan(True)+++if __name__ == "__main__":+ unittest.main()diff --git a/.agents/skills/verify-openhands/scripts/lib/websocket-events.mjs b/.agents/skills/verify-openhands/scripts/lib/websocket-events.mjsnew file mode 100644index 0000000000..ea84e2ddc9--- /dev/null+++ b/.agents/skills/verify-openhands/scripts/lib/websocket-events.mjs@@ -0,0 +1,67 @@+// Subscription metadata only: payloads and headers never enter the event log.+import { redactUrl } from "./network-bodies.mjs";++export function webSocketMetadata(raw) {+ try {+ const url = new URL(raw);+ if (url.href === "about:blank") return { url: url.href };+ if (!["http:", "https:", "ws:", "wss:"].includes(url.protocol))+ return { url: "<unavailable>" };+ url.username = "";+ url.password = "";+ url.hash = "";+ const cursors = url.searchParams.getAll("after_seq");+ const cursor = cursors.length === 1 ? cursors[0] : undefined;+ const afterSeq = /^-?\d+$/.test(cursor ?? "") ? cursor : undefined;+ // Allow only one numeric cursor. Even unknown parameter names/values+ // may contain credentials, so omit the rest rather than guessing.+ const redactedQueryParameters =+ [...url.searchParams].length - (afterSeq === undefined ? 0 : 1);+ url.search = "";+ if (afterSeq !== undefined) url.searchParams.set("after_seq", afterSeq);+ return {+ url: redactUrl(url.toString()),+ ...(afterSeq === undefined ? {} : { afterSeq }),+ ...(redactedQueryParameters ? { redactedQueryParameters } : {}),+ };+ } catch {+ return { url: "<unavailable>" };+ }+}++export function isWebSocketObservation(kind) {+ return ["websocket-created", "websocket-open", "websocket-close"].includes(+ kind,+ );+}++export function observeWebSocket(socket, record, { connectionId, pageUrl }) {+ const metadata = webSocketMetadata(socket.url());+ let receivedFrames = 0;+ let sentFrames = 0;+ let observedOpen = false;+ const emit = (kind, detail = {}) =>+ record(kind, {+ connectionId,+ ...metadata,+ page: webSocketMetadata(pageUrl()).url,+ receivedFrames,+ sentFrames,+ ...detail,+ });+ emit("websocket-created");+ const frame = (received) => {+ if (received) receivedFrames += 1;+ else sentFrames += 1;+ if (!observedOpen) {+ observedOpen = true;+ // Playwright exposes creation, frames and close, but no handshake-open+ // event. The first frame proves communication; this is its timestamp.+ emit("websocket-open", { observed: "first-frame" });+ }+ };+ socket.on("framereceived", () => frame(true));+ socket.on("framesent", () => frame(false));+ socket.on("socketerror", () => emit("websocket-error"));+ socket.on("close", () => emit("websocket-close"));+}