Garry's Opinionated OpenClaw/Hermes Agent Brain
TypeScript
Merge remote-tracking branch 'origin/master' into capy/gbra175-autoupgrade-serve-restart
garrytan4641f1efbd5a
diff --cc docs/architecture/key-files/commands-6.mdindex f63c1a1ed6,e4d832f10c..a1475dc70b--- a/docs/architecture/key-files/commands-6.md+++ b/docs/architecture/key-files/commands-6.md@@@ -13,8 -13,8 +13,8 @@@ Current behavior and load-bearing invar- `src/commands/transcripts.ts` — the transcripts command family, all local-only (`ctx.remote=false` by construction). `recent`: raw `.txt` corpus reads via `listRecentTranscripts` (same library as the gated `get_recent_transcripts` op). `ingest <path-or-glob>`: the cross-harness session importer — resolves ONE source id (6-tier chain), threads activePack once, streams progress (phase `transcripts.ingest`), and calls `runTranscriptsIngest` with the CLI's `makeContext` context, so on a writer-claimed brain each rendered part is submitted as a `put_page` write request naming the page's current revision (a part whose publication would change nothing is screened request-free first and takes no admission; unmanaged brains and dry runs keep the direct importer; `recover codex --apply` passes the same context; pinned by `test/transcripts-managed-ingest.test.ts`); embedding is OFF by default (embed backfill is the catch-up lane; the embed flag opts in); the max-bytes flag (validated size string, e.g. `4gb`) overrides the per-format file/store byte caps for oversized stores while omission preserves each adapter's native default; the since-last watermark is an op-checkpoint (op `transcripts-ingest`, fingerprint = source + pathspec + format + adapter version + any explicit byte cap via `ingestCheckpointFingerprintInput` — a checkpoint written under one cap is never silently reused under another, so a capped run's skipped tail can't read as scanned; the pathspec binds the user-stated paths resolved + sorted, and the no-arg/all lane substitutes hostname + sorted harness roots so DB-backed checkpoints shared across a brain's machines never let one machine inherit another's watermark and skip sessions it never scanned) advanced ONLY after a clean, untruncated, non-dry scan (a Codex sub-agent rollout, `session_meta` payload.source.subagent, has no user turn by design: the adapter reports `subagentRollout`, the scan counts it in `subagentRollouts` and prints `N sub-agent rollouts (no user turns by design)`, and it keeps the scan clean; an ordinary assistant-only rollout is still `userTurnsMissing` drift); no-arg = confined discovery table, the all flag imports the discovered set. `status`: found-vs-imported gap table (disk scan vs ONE conversation-pages frontmatter query — `executeRaw`, both engines — in `src/core/transcripts/discover.ts`) — the correctness surface that catches late-arriving sessions no watermark can. The facts flag hands EVERY touched slug (including hash-skipped) to `src/core/transcripts/ingest-facts.ts`; a page that fails extraction counts in `facts.pagesFailed`, stays unfinished, and makes the command exit 1. Registered in CLI_ONLY + CLI_ONLY_SELF_HELP + SELF_HELP_WITHOUT_ENGINE (engine-free help). `expandTilde(raw)` (bare `~`, leading `~/` or `~\\` → home dir; `~user` forms and a mid-string `~` stay literal) is the one tilde expansion `expandPaths` applies to every path spec. `recover codex` (`src/commands/transcripts-recover.ts`, logic `src/core/transcripts/recover.ts`; D15 for #5163): finds codex conversation pages in one source with no `**User**` anchor, re-parses the retained rollouts (discovery root + archived store, or given paths) and classifies each session recoverable / still userless / unrecoverable (rollout gone); preview by default, `--apply` re-imports only the recoverable rollouts through `runTranscriptsIngest` (content-hash dedup, no facts, no watermark write). Pinned by `test/transcripts-recover-codex.test.ts`. `audit-secrets [--source-id S] [--json]` (`src/commands/transcripts-audit.ts`, logic `src/core/transcripts/secret-audit.ts`): the read-only installed-base audit that streams imported conversation pages (`type = 'conversation'` with `transcript_import` frontmatter) in id-keyset batches of 25, scans title and body with the transcript-lane detectors, lists slug, source, hit counts per pattern and up to 20 line numbers (never values or previews), and caches the summary without slugs in config row `transcripts.secret_audit` for the doctor check `transcript_secret_exposure` (`src/commands/doctor/checks/transcript-secrets.ts`, brain category), which counts imported conversation pages and reads that summary (never scans): no audit, an older `TRANSCRIPT_SECRET_AUDIT_VERSION` or a source-scoped audit warns with the audit as fix; hits warn with the audit's age; a clean full audit is ok. Pinned by `test/transcripts-secret-audit.test.ts`.- `src/commands/upgrade.ts` — self-update CLI. It sets `GBRAIN_UPGRADE_OWNS_MIGRATIONS=1` for the install step, so package postinstall (`scripts/postinstall.ts`) skips `apply-migrations`; post-upgrade owns them. An install step that fails or times out after `gbrain --version` already reports a newer version counts as installed. The run ends with a `Binary:` line (installed version, or upgrade failed) and a `Migrations:` line: complete; running, with the lock holder's host and pid when post-upgrade exits 75; or failed, with the recovery command (`--force-retry <v>` for wedged versions). `runPostUpgrade()` enumerates migrations from the TS registry (`src/commands/migrations/index.ts`) and tail-calls `runApplyMigrations(['--yes', '--non-interactive'])` so the mechanical side of every outstanding migration runs unconditionally. After schema apply it prints the one-shot auto_chronicle default-on `[AGENT]` notice (`core/chronicle/upgrade-notice.ts`). Before the swap of a bun-link, package or ClawHub install it refuses (exit 78, nothing changed) a target whose Bun floor is above the host's Bun or unreadable (`--no-bun-floor-check` overrides, manual only; `src/core/bun-floor.ts`); after the swap it runs `gbrain --help` as a smoke that passes the startup Bun check and, on failure, exits 1 with per-method recovery (#5855). Everything the post-swap tail needs is imported statically, so the old process never loads new-release modules into its cached graph (#6604; the supervisor step constant lives in the import-free leaf `src/commands/movement-step.ts`; guard: `test/upgrade-post-swap-imports.test.ts`). A tag-pinned reinstall hint (`bunReinstallCommand`, `reinstallHintLines`) probes `git ls-remote --exit-code --tags … refs/tags/v<x>` before `bun remove -g gbrain`, prints a restore line, and on Windows prints one step per line; `planPinnedReinstall` probes the tag itself (published / not published / unknown) and falls back to the newest published `v<x>` tag (`latestPublishedVersion`) when the requested one is missing (#6662, D24; `test/upgrade-release-tag-hint.test.ts`). A verified version equal to the running one after the install step is no swap (#6828): it prints `No newer version was installed`, skips `last_upgrade`, `JUST_UPGRADED`, the cache reset and the smoke, and on a tag pin with no newer published tag the hint moves the install off the pin with `to_version: latest`, never a cached unpublished version (#6718). With `GBRAIN_SWAP_RECEIPT` set it writes a `SwapReceipt` (`installed` / `not_installed` / `installed_broken`) for the autopilot channel (`test/autopilot-unpublished-target-6828.test.ts`). Tests: `upgrade-migration-outcome.test.ts`, `upgrade-no-autopilot.serial.test.ts`, `upgrade-bun-link-arc.serial.test.ts`. `post-upgrade --json` (D2, `json: 'document'`): human lines go to stderr and stdout is one report `{status, apply_migrations:{exit_code}, schema, warnings[]}`; it calls `applyMigrations()` and still ends on any status apply-migrations returns (a failure is one envelope leading with the report). GBRA-175: `--channel latest|stable` (default `self_upgrade.channel`): `resolveStableRelease` maps the commit `refs/tags/stable` peels to onto its `v<version>` tag through `git ls-remote`, and `installStableRelease` installs exactly that release (bun: remove then add `#v<version>`, restoring the previous release on failure; bun-link: fast-forward only; binary: `fetchReleaseByTag`), never downgrading. Before the swap `preflightHostRestart` refuses (`restart_refused`, exit 3) when a running process could not be restarted; after post-upgrade the new build runs `gbrain upgrade --restart-only` (`--no-restart` skips it). Tests: `upgrade-stable-channel.test.ts`. A global install pinned to a tag or commit (`bunGlobalExactTagPin(root, true)`) is replaced with the newest published release the same way instead of `bun update`, which keeps the pin. `parseChannelArg` and `resolveStableRelease` live in `src/core/stable-channel.ts` (re-exported).-- `src/commands/upgrade-restart.ts` — GBRA-175: the host side of the upgrade restart. Real process sources (consumer heartbeat rows on Postgres, the supervisor PID file, the worker registry, `ps -U <uid> -ww -o pid=,ppid=,etime=,command=` with `/proc/<pid>/environ` GBRAIN_HOME matching on Linux; the install version above each entry point), the installed serve and autopilot services with the pid each manager runs (`hostServices`: `systemctl --user show -p MainPID`, `launchctl list`), `terminateIfUnchanged` (SIGTERM an MCP pid only while `ps` still shows the same server and start time), `restartRefusal` (ask_user + `--stop-mcp` when only MCP servers block, else the first manual step for the user), `preflightHostRestart` (pre-swap, every process stale) and `runHostRestart` (`gbrain upgrade --restart-only [--stop-mcp] [--dry-run] [--json]`: plan, run, verify for up to 180 s with the warm-socket probe `probeCliSocket`). Tests: `upgrade-restart.test.ts`.+- `src/commands/upgrade-restart.ts` — GBRA-175: the host side of the upgrade restart. Real process sources (consumer heartbeat rows on Postgres, the supervisor PID file, the worker registry, `ps -U <uid> -ww -o pid=,ppid=,etime=,command=` with `/proc/<pid>/environ` GBRAIN_HOME matching on Linux; the install version above each entry point), the installed serve and autopilot services with the pid each manager runs (`hostServices`: `systemctl --user show -p MainPID`, `launchctl list`), `terminateIfUnchanged` (SIGTERM an MCP pid only while `ps` still shows the same server and start time), `restartRefusal` (ask_user + `--stop-mcp` when only MCP servers block, else the first manual step for the user), `preflightHostRestart` (pre-swap, every process stale) and `runHostRestart` (`gbrain upgrade --restart-only [--stop-mcp] [--dry-run] [--json]`: plan, run, verify for up to 180 s with the warm-socket probe `probeCliSocket`). #7475: `restartServeAfterSelfUpgrade`, called each autopilot-daemon tick, restarts the serve the serve service runs once it is on an older build than the relaunched autopilot. It runs once per autopilot process, only in auto mode, in quiet hours and when the brain is idle (`computeAutopilotIdle`), and audits as `restart_serve`. `planServeRestart` reports a serve the service does not run, or one on another install, as `manual` and never restarts it. Tests: `upgrade-restart.test.ts`, `autopilot-serve-restart-7475.test.ts`.- - `src/core/upgrade-restart.ts` — GBRA-175: the pure half of the upgrade restart. `collectHostProcesses` (the process table is the authority: a heartbeat, PID-file or registry record whose pid `ps` does not show as gbrain is dropped; `gbrainEntry` skips wrappers such as tini, and a process's parent is its nearest gbrain ancestor), `planRestart` (a service restarts only processes under its manager's pid; `supervisorRestartArgv` keeps the launch options and PID file; anything else is a `ManualStep`), `runRestartSteps`, `restartProblems` (old pids gone, one build, roles and the supervised worker back, warm socket accepts) and `unverifiedBuilds`. A stale process whose `install_root` is another install than the one being upgraded gets a `replace_copy` step first when it is a Bun global copy (`bunGlobalCopyOf`: probe the tag, `bun remove -g` then `bun add -g` under its `BUN_INSTALL`, restore on failure), else a manual step from `copyFix`; after the restart `mismatchedBuilds` names any process back on another build, and the host's `mismatchFix` chains that replacement with `gbrain upgrade --restart-only`.+ - `src/core/upgrade-restart.ts` — GBRA-175: the pure half of the upgrade restart. `collectHostProcesses` (the process table is the authority: a heartbeat, PID-file or registry record whose pid `ps` does not show as gbrain is dropped; `gbrainEntry` skips wrappers such as tini, and a process's parent is its nearest gbrain ancestor), `planRestart` (a service restarts only processes under its manager's pid; `supervisorRestartArgv` keeps the launch options and PID file, read from `--pid-file <path>`, and drops a `--flag=value` token that is neither a global flag nor `--job-isolation=`, because the supervisor that holds it ignored it and this build's `=` gate refuses it (#7619); anything else is a `ManualStep`), `runRestartSteps`, `restartProblems` (old pids gone, one build, roles and the supervised worker back, warm socket accepts) and `unverifiedBuilds`. A stale process whose `install_root` is another install than the one being upgraded gets a `replace_copy` step first when it is a Bun global copy (`bunGlobalCopyOf`: probe the tag, `bun remove -g` then `bun add -g` under its `BUN_INSTALL`, restore on failure), else a manual step from `copyFix`; after the restart `mismatchedBuilds` names any process back on another build, and the host's `mismatchFix` chains that replacement with `gbrain upgrade --restart-only`.- `src/commands/watch.ts` — `gbrain watch`: the push transport. Reads turns from stdin as they arrive (`user:`/`assistant:` prefixes; unprefixed = user turn), keeps a rolling in-process window (`--window-turns`, default 4), calls `volunteerContext` per turn, streams pointers to stdout (`--json` for JSONL with turn attribution), logs `channel: 'watch'` events with a per-session id. Session dedupe feeds already-pushed slugs back as priorContext so the core's slug-only suppression dedupes. Blocks in the stdin iteration (interactive alive until Ctrl-C/Ctrl-D; piped ends at EOF) — deliberately NOT in DAEMON_COMMANDS; SIGINT closes the stream so teardown flows through finishCliTeardown. Per-turn resolution failures are fail-open. Registered in CLI_ONLY + CLI_ONLY_SELF_HELP + THIN_CLIENT_REFUSED_COMMANDS (thin clients use the `volunteer_context` MCP op). Pinned by `test/watch-command.test.ts`.diff --cc docs/guides/upgrades-auto-update.mdindex c7b2018135,409d6dfe8e..3668ca2b73--- a/docs/guides/upgrades-auto-update.md+++ b/docs/guides/upgrades-auto-update.md@@@ -275,18 -275,14 +275,23 @@@ gbrain upgrade --restart-onlgbrain upgrade --no-restart # install only; doctor fails runtime_build_skew until the restart runs```+ On an always-on host, start with `gbrain autopilot pause --reason "upgrade"`+ and wait until `gbrain jobs list --status active --json` returns no jobs, so no+ job runs half on each build; resume with `gbrain autopilot resume` once the+ checks below pass ([the order](../../skills/gbrain-upgrade/SKILL.md#always-on-hosts-pause-drain-upgrade-verify-resume)).+-The autopilot `--swap-only` channel still exits for its own supervisor to-relaunch and restarts nothing else; doctor's `runtime_build_skew` fails while-two builds run, with `gbrain upgrade --restart-only` as its fix.+The autopilot `--swap-only` channel exits for its own supervisor to relaunch.+Once the relaunched autopilot runs the new build, it checks the serve that the+installed serve service runs (`com.gbrain.serve` under launchd,+`gbrain-serve.service` under systemd). If that serve is on an older build, the+autopilot restarts it through the service and verifies that it came back on+the new build. It does this once per autopilot process, under the swap's own+auto-mode, quiet-hours and idle gates, and records it in the self-upgrade+audit as `restart_serve` (`applied`, or `skipped` with `manual` or+`unverified`). It leaves alone a serve that the service does not run (one+started by hand or from another install), the jobs supervisor and MCP+servers. For those, doctor's `runtime_build_skew` fails while two builds run,+and `gbrain upgrade --restart-only` is its fix.<a id="restart-and-verify"></a>### Restart and verify: a deploy ends with data movementdiff --cc src/core/behavior-change-notice.tsindex 59d3361c5c,4150de8f58..add1e2341a--- a/src/core/behavior-change-notice.ts+++ b/src/core/behavior-change-notice.ts@@@ -1223,8 -1248,124 +1248,125 @@@ export const BEHAVIOR_CHANGES: Readonly{ since: '0.60.427.0', text: 'More gbrain state follows GBRAIN_HOME: feature offers (`feature-offers.json`), the upgrade state and error log, `mounts.json` and the mounts cache, the brain-pull log (including the log paths a newly installed launchd pull job writes), the default supervisor pid file, skillopt captures, the transcript harvest pattern file, notability eval files and the greenfield import audit directory now live under `<GBRAIN_HOME>/.gbrain` when GBRAIN_HOME is set. Before, they were written under `$HOME/.gbrain` whatever GBRAIN_HOME said. With GBRAIN_HOME unset nothing moves. A host that sets GBRAIN_HOME keeps its old copies under `$HOME/.gbrain`: move them by hand if you want the old offers, mounts or upgrade history, and restart a supervisor that was started before the upgrade so `gbrain jobs supervisor status` finds its pid file.' },// GBRA-175 (stable channel, upgrade restarts every process). `since` is stamped at the merge slot.{ since: '0.60.428.0', id: 'upgrade_restarts_7245', text: '`gbrain upgrade` now restarts every gbrain process this brain runs on the host onto the new build (serve and autopilot through their installed services, the jobs supervisor and its worker with `gbrain jobs supervisor stop` then `start --detach`) and verifies that they all report one build and that the warm CLI socket accepts it. Before installing anything it refuses with `restart_refused` (exit 3) when a process cannot be restarted automatically: a serve or worker started by hand, an autopilot without its service, a running sync, or an MCP server an agent harness started (rerun with `--stop-mcp` to stop those; reconnect gbrain in each harness afterwards). `--no-restart` installs without restarting; `gbrain upgrade --restart-only` restarts onto the installed build (`--dry-run` previews). The autopilot `--swap-only` channel is unchanged.' },+ { since: 'TBD', id: 'autopilot_restarts_serve_7475', text: 'With `self_upgrade.mode auto`, the autopilot that relaunches after a self-upgrade now restarts the serve your installed serve service runs (`com.gbrain.serve` under launchd, `gbrain-serve.service` under systemd) when that serve is still on the old build, and verifies it came back on the new one. It runs once per relaunch, in quiet hours, when the brain is idle, and audits as `restart_serve`. A serve started by hand or from another install is never restarted: doctor `runtime_build_skew` keeps failing with `gbrain upgrade --restart-only` as its fix.' },{ since: '0.60.428.0', id: 'runtime_build_skew_fails_7245', text: 'doctor `runtime_build_skew` now fails (was a warning) when two or more gbrain builds run on this host, and its fix asks the user before running `gbrain upgrade --restart-only`; one process older than its install alone still warns. New release channel: `gbrain upgrade --channel stable` or `gbrain config set self_upgrade.channel stable` installs only releases a maintainer promoted after the production-shaped soak passed on their exact tree, and the update checks then offer only that release; the default channel (`latest`) is unchanged.' },+ { since: '0.60.430.0', text: 'In a managed canonical worktree, `gbrain git commit-push -m <msg> -- <paths>` now commits the named paths there (outcome `committed_local`, exit 0) instead of refusing `writer_coordinator_required`. It takes the commit lock and commits only those paths: it never fetches, merges, pushes or moves the checkout, so the persistence owner stays the only publisher; the owner\'s next Git publish, or `gbrain git commit-push --push-only --keep-checkout`, pushes the commit. --pending, --branch, --resolve, --expect-remote and --allow-outside-source-root are refused there (`commit_paths_invalid`).' },+ { since: '0.60.430.0', text: 'The `refresh_diverged` refusal of `gbrain sources refresh` now counts the checkout\'s local commits and its fix is `gbrain git commit-push --push-only --keep-checkout --path <root> --json` (actor user, ask_user): it merges the upstream in memory, refuses on a conflict, pushes the merge and never moves the checkout, after which the refresh fast-forwards. It used to suggest reconciling in a separate clone or recloning, which drops the local commits.' },+ { since: '0.60.430.0', text: '`gbrain sources writer status --json` has a new `checkouts` block: each canonical checkout on this host against its last fetched upstream (no fetch), with branch, upstream, ahead, behind, `diverged`, each commit the upstream lacks (sha, author, date, subject, paths; at most 20) and porcelain counts (staged, modified, untracked, conflicted). doctor `git_convergence` adds the same commits and counts for a managed canonical checkout, and now warns at once when one has diverged from its upstream (it blocks `gbrain sources refresh`), with the keep-checkout publish as an ask-user fix.' },+ { since: '0.60.430.0', text: 'New `gbrain sources set-sync <id> <on|off>` sets `config.syncEnabled`, which includes or excludes a source from automatic sync (`sync --all`, autopilot\'s freshness dispatcher and the full-cycle fan-out; an explicit `gbrain sync --source <id>` runs either way). `--dry-run` prints the current state and the apply command, which carries `--expect <current>`; an apply whose expectation no longer holds refuses `preview_changed` and changes nothing. Setting it used to need a raw write to `sources.config`.' },+ { since: '0.60.430.0', text: 'New doctor check `cron_gbrain_home` reads `crontab -l` (read-only, 5 s; no result unless the crontab mentions gbrain or src/cli.ts) and the scripts its lines run (at most 20, 64 KiB each) and warns on each command that runs `src/cli.ts` directly without GBRAIN_HOME, or the `gbrain` binary without GBRAIN_HOME while this brain runs under one: cron passes no GBRAIN_HOME, so those commands act on `~/.gbrain` instead of this brain. A GBRAIN_HOME assignment on the line, in the crontab\'s environment lines or in the script counts. The fix asks the user to edit the crontab.' },+ // GBRA-190 (verified-bug wave). `since` is stamped at the merge slot.+ { since: '0.60.433.0', text: 'Upgrading builds the index `pages_source_created_idx` (migration v270) over live pages by source and creation time: on Postgres online (`CREATE INDEX CONCURRENTLY`; reads and writes go on while it builds), on PGLite as a plain build that holds writes until it ends. With it, `gbrain facts relink --conversation` (dry runs included) and doctor `unlinked_facts` check for a page created since each relink attempt with one index probe instead of reading every live page of the source, which ran into `statement_timeout` on large brains (#7379).' },+ { since: '0.60.433.0', text: 'A `remember` items batch now applies the top-level `entity`, `content_origin` and `valid_from` to every item that does not set its own, as it already did for `provenance`, `kind`, `ttl` and `visibility`, and accepts `content_origin` and `valid_from` on an item. Before, a top-level `entity` was ignored (the items were saved with no subject and a `NO_ENTITY` warning), a top-level `content_origin: tool_output` was ignored, so batched tool output was stored at the writer\'s trust tier instead of `external_untrusted`, and an item carrying `content_origin` or `valid_from` was refused (#7432).' },+ { since: '0.60.433.0', text: 'Conversation-facts extraction leaves out an extracted claim that matches a purge tombstone and publishes the rest of the page (the run reports `facts_skipped_purged`). Before, that one claim made the facts insert guard refuse the page\'s whole insert after the model call was paid for, and on a managed brain every page in the same batch request failed with it and was blocked after three attempts until it changed (#7438).' },+ { since: '0.60.433.0', text: 'The extract_atoms backlog (doctor, `dream --phase extract_atoms --drain` and autopilot drains) now counts a page whose last extraction was cut at a lower `cycle.extract_atoms.max_input_chars` than the current cap, so raising the cap re-extracts those pages, with model calls, instead of the drain reporting drained with 0 batches. At an unchanged or lowered cap nothing changes (#7391).' },+ { since: '0.60.433.0', text: 'Hermes transcript import keeps the answers a user gave in a `clarify` form (answered responses of a submitted form, or the answers a timed-out form had already locked) as user turns, each after its question as the assistant\'s turn. A Hermes session imported before re-renders on its next import scan when it holds such answers, which can run fact extraction for that page again (#7396).' },+ { since: '0.60.433.0', text: 'Fact subject inference links more facts and fewer wrong ones: a currency code next to an amount (`CHF 80`) and a record label opening a sentence (`Invoice ...`, `Work session ...`) no longer block a fact that names one person or company, and an entity introduced only as an example (`including X notes`, `for example, X`) is never chosen as the subject. A page named like the code or label still competes. A sentence that opens with `For example, X ...` no longer takes X as its subject (#7405).' },+ { since: '0.60.433.0', text: '`gbrain think` gives reasoning models from gpt-5 on 16,000 output tokens behind any provider or router prefix, as it already did for gpt-5 and the o-series on their own provider; gpt-6 ids get it too. Before, those got 4,000 and long answers came back truncated (`LLM_OUTPUT_TRUNCATED`). Providers bill the tokens used, not the cap (#7421).' },+ { since: '0.60.435.0', text: 'A search for a full name with a short first name ("Matt Example") now finds the page titled "Matthew Example", and back (Bob and Robert, Liz and Elizabeth and the other common given-name groups): the title arm also matches the other forms of the given name, and the title boost counts them as a title match. When no title holds every word of a query of up to four words, the title arm\'s any-word retry now puts titles holding the rarest word first instead of tying them with titles that hold only a common word (#7471).' },+ { since: '0.60.435.0', id: 'keyword_or_rarity', text: 'When no page holds every word of a keyword search of up to four words, the any-word retry of the keyword arm now ranks pages holding the rarest query words first, instead of ranking pages that repeat a common word above the few that hold a rare one. The strict all-words search and longer queries are unchanged.' },+ // GBRA-156 (exact identity lookup, same release as the #7471 row). `since` is stamped at the merge slot.+ { since: '0.60.435.0', id: 'whois_exact_identity', text: 'New read operation `whois` (CLI `gbrain whois <email or name>`, MCP tool `whois`): an exact identity lookup for agents triaging email or calendar. It returns the entity pages whose frontmatter `email`/`emails`, title or declared alias equals the query, ignoring case, accents and surrounding quotes, labels each match `email`, `title` or `alias`, sets `ambiguous` when more than one page matches, and returns an empty list, never a search fallback, when none does. Remote callers see only their granted sources and no private or quarantined page. Nothing existing changes.' },+ // GBRA-190 connectors wave. `since` is stamped at the merge slot.+ { since: '0.60.437.0', text: 'Chat-connector syncs stop instead of retrying through a block: a bot challenge on any status (the `cf-mitigated: challenge` header, a challenge page on a 5xx, a redirect to a Cloudflare challenge) now ends the run as `forbidden`; a 429 that outlasts its four retries, or asks for a `Retry-After` over two minutes, ends it as the new status `rate_limited` without fetching the remaining conversations; and five conversation fetches in a row failing with a server error or no response stop it as `partial` with `stopReason: consecutive_fetch_failures`. The next sync resumes from the same place (#7482).' },+ { since: '0.60.437.0', text: 'A chat-connector cookie or token holding a line break, NUL, another control character or a non-ASCII character is refused before any request (`invalid_header_value`): `connectors auth` reports it and `connectors sync` returns `auth_required` with that hint and stamps the auth error. A request that fails before a response is reported by its error code (`ECONNREFUSED`, `TypeError`), never by the runtime\'s message, so a header value can never be echoed into probe output, sync hints or logs (#7483).' },+ // GBRA-133 G2 (superseded remote blobs). `since` is stamped at the merge slot.+ { since: '0.60.438.0', text: '`gbrain git commit-push` now keeps this machine\'s version of a file the remote changed back to a version this checkout already had at that path and later replaced (a stale clone or writer pushing an old copy, often inside a merge), instead of stopping with `integrate_conflict`. It reads the last 200 commits that changed the file. The push lists those files in `superseded_remote_paths` and the merge message names them. A version this checkout never had, a remote deletion and a file named with `--resolve` still conflict; when both kinds meet, `integrate_conflict` lists only the real conflicts and `detail.superseded` names the others.' },+ // GBRA-60 G5 (authored writes get a markdown form). `since` is stamped at the merge slot.+ { since: '0.60.441.0', text: 'On a managed brain, `add_link` and `remove_link` now edit a `## Links` table in the page\'s Markdown (one row per target, type and source, between `<!--- gbrain:links:begin -->` and `end` markers) through the page-mutation door, and the edge in the database is derived from that row, so a manual edge survives a rebuild from Git. Calling `add_link` again on an existing row without `context` or dates keeps the stored ones. Manual edges recorded before this release stay in the database as they were. Unmanaged brains are unchanged.' },+ { since: '0.60.441.0', text: 'On a managed brain, `remove_link` no longer deletes an edge that the page\'s own text creates (a link, wikilink or frontmatter field): it keeps it and returns `derived_from_text` with a message saying to edit the page text (`edit_page`) instead.' },+ { since: '0.60.441.0', text: 'On a managed brain, `extract_entities` writes the entity\'s timeline bullet into the entity page\'s Markdown through the page door, and the backlink to the source page is the link inside that bullet rather than a database-only edge.' },+ // GBRA-197 lane A (#7514). `since` is stamped at the merge slot.+ { since: '0.60.443.0', text: 'On a managed brain, an atom batch whose write did not commit is now reported as a write failure: each item in `gbrain dream --drain --json` carries the request id (`gbrain write-request -- <id>` explains it) and the refusal code, and a batch of such items no longer stops the drain as `provider_failure` (code `atom_drain_provider_failure`, `transient`, retryable). After 3 such failures on the same content the page or transcript is held (`held_failed`, doctor `extract_atoms_held_failed` with its receipt and retry command) instead of failing every window; no model call is made for it, and an explicit retry runs it again. An extraction whose model reply repeats an atom title now publishes one atom for that title instead of failing the whole batch.' },+ // #7516 (GBRA-197 lane B). `since` is stamped at the merge slot.+ { since: '0.60.443.0', text: 'A managed conversation-facts batch whose receipt lost its stored intent (a `purged_content` refusal the journal redacts, a purge, or receipt compaction) no longer sticks its pages. `gbrain extract-conversation-facts` derived the same batch request id again, so every run paid a model call, failed with `idempotency_conflict` (or got the old failure back) and inserted nothing. The batch now publishes under its pages\' next attempt after one more model call, and the next run makes none. A crashed run that resubmits the same batch still replays its receipt.' },+ // GBRA-197 (#7515 chat-connector hardening). `since` is stamped at the merge slot.+ { since: '0.60.443.0', text: 'Chat-connector syncs no longer refresh and retry after a 403: a 403 with a JSON error body or a `www-authenticate` header ends the run as `auth_required` at once, and a redirect is never followed (the session cookie only goes to the provider\'s own origin) and ends it as `auth_required` too. ChatGPT\'s token refresh now goes through the same client, so it is paced, counted and refused the same way: a refused refresh reports `forbidden`, `rate_limited` or `auth_required` instead of being ignored, and `gbrain connectors auth` stops there instead of probing. Re-copy the cookie when a sync reports `auth_required`.' },+ { since: '0.60.443.0', text: '`gbrain connectors sync --all` now syncs only providers whose `connectors.<p>.auto_sync` is on and names each one it skips; a queued `connector-sync` job rechecks `auto_sync` when it starts and returns `skipped` if it was turned off. A host cron running `connectors sync --all` for a provider without `auto_sync` stops syncing it: turn `auto_sync` on, or run `gbrain connectors sync <provider>`, which still syncs either way.' },+ { since: '0.60.443.0', text: 'Every chat-connector sync is bounded: at most 10,000 requests per run (5,000 per Claude organization, and at most 10 Claude organizations listed), list pages cut to the page size asked for, and new `--max-requests N` and `--deadline <90s|30m|2h>` flags on `gbrain connectors sync`. A run that reaches a bound ends `partial` with `stopReason` `max_requests` or `deadline`, keeps what it imported and holds the watermark; run `gbrain connectors sync <provider>` again (without `--full`) to continue.' },+ { since: '0.60.444.0', text: '`gbrain git commit-push` takes `-m` more than once, like `git commit`: each `-m` is a new paragraph of the message (a second `-m` carries a trailer). With `--path <dir>` below the checkout top, the paths after `--` are now relative to `<dir>` (like `git -C`), as its help said; they used to be read relative to the checkout top. The commit step that `gbrain repair frontmatter`, `gbrain repair content`, `gbrain repair file-drift` and the legacy fence repair (`gbrain sources status`, doctor `fence_integrity`) print for a legacy source is now `gbrain git commit-push --path <root> -m <subject> [-m <trailer>] -- <paths>`, which takes the writer lock, commits only those paths, merges remote commits and pushes, instead of a bare `git add` plus `git commit` (the frontmatter step named no paths on its commit, so it also committed anything else staged). In a checkout with no Git remote at all (none configured and the branch names none), `gbrain git commit-push -m <msg> -- <paths>` now commits the named paths locally under the commit lock (outcome `committed_local` with `no_remote: true`, exit 0) and says nothing was pushed; it used to refuse `git_unavailable` and commit nothing. `--push-only`, `--pending`, `--branch`, `--resolve`/`--expect-remote` and `--require-clean` are refused there (`commit_paths_invalid`), and a branch that names a missing remote still refuses `git_unavailable`.' },+ // GBRA-60 G6 (strict git-canonical sources are door-only). `since` is stamped at the merge slot.+ { since: '0.60.445.0', text: 'On a brain whose writes are not managed, a direct import into a source configured `canonical: git` (`gbrain import` of markdown, code or an image, the streamed large-page import, and `extract_entities` creating or updating entity pages) is now refused with `strict_requires_managed` before anything is written, because only the managed writer commits a git-canonical source through its one door. Other sources and managed brains are unchanged.' },+ { since: '0.60.445.0', text: 'On a managed brain, maintenance writers publish one commit per run instead of one per page: a dream patterns or concept republish batch, the entity stubs and timeline bullets of one `extract_entities` call, and one atom generation (its new atoms, activation and retirements) each publish in one transaction and one git commit. Each page still keeps its own request, receipt and failure; a page that fails no longer holds back the rest. `brainstorm --save` on a managed brain now reports its page as accepted and pending (exit 10) with the receipt to poll.' },+ { since: '0.60.445.0', text: 'On a source configured `canonical: git`, a dream synthesis page is checked against its transcript and stamped with provenance before it is written, so a quote the model invented never reaches git history, and a world concept its provenance covers is published once at world. A connector page marked private, and a private calendar event, is stored database-only and never committed.' },+ // GBRA-95 G7a (declared database-only classes). `since` is stamped at the merge slot.+ { since: '0.60.448.0', id: 'git_canonical_declared_classes', text: '`gbrain sources git-gap` now declares every page of a source without a checkout `database_only_source` instead of counting it as not compared. On a git-canonical source (`config.canonical = "git"`, off by default and set by nothing yet), private, alias-private and lineage-private pages, `storage.db_only` directories, never-filed derive folders and ignored paths keep no file in Git (no file, no commit, no Git effect), a page turning private is published as the commit that removes its file, and a write that would commit a private page (`git_private_commit_refused`), store an undeclared page without a file (`git_home_required`), commit a file under an LFS rule without `storage.git_lfs` (`git_lfs_not_enabled`) or come from a writer without the git protocol (`git_protocol_required`) is refused.' },+ { since: '0.60.452.0', id: 'starter_tools_list_params_gbra156', text: 'The starter MCP tool list is about 27% smaller: it leaves out parameters MCP callers cannot use (local-only, server-set provenance) and rarely needed ones (legacy, admin, versioned skill catalog, image and code search, tuning and recovery knobs with safe defaults). Every left-out parameter is still accepted, and `request_tools {tools: [names]}` now returns each tool\'s full schema with every optional parameter, also on a server capped at starter. Memory verb descriptions no longer start with `MEMORY VERB (v1):`. The verbs and full surfaces list every parameter as before.' },+ { since: '0.60.452.0', id: 'sources_not_readable_gbra156', text: 'An unscoped `search`, `query`, `recall` or `list_pages` from a token or agent connection with no source grant that comes back empty, while every source it can read holds no page and other sources do, now carries a `sources_not_readable` notice (kind degraded). It says the empty result is not proof of absence, never names a hidden source, and its fix is `gbrain sources federate <source>` on the brain host, or a grant for the token. Granted tokens and OAuth clients never get it.' },+ // GBRA-199. since: '0.60.503.0' is stamped at the merge slot.+ { since: '0.60.503.0', id: 'registration_lists_verbs_gbra199', text: 'New MCP registrations gbrain writes now list only the seven memory verbs (`recall`, `remember`, `entity`, …) and `request_tools` in tools/list: `gbrain setup claude-code|codex`, `gbrain bootstrap hooks`, the readiness and `gbrain init` registration lines (`serve --surface starter --advertise verbs`: the starter tools stay callable) and `gbrain mcp expose` (its `serve --http` runs with `--surface full --advertise verbs`, so clients without their own surface see that list and every operation stays callable). `request_tools` lists the rest and returns any tool\'s full schema, and a tool it describes joins the session\'s list. The initialize instructions follow the listed tools, so they point agents at `recall` instead of `search`. Existing registrations and exposures keep their listing, and an explicit `--surface full` (or starter, or verbs) lists that whole surface. New `gbrain serve --advertise <verbs|starter|full>` sets the listing per server and wins over config `mcp.advertised_surface`; a listing narrower than the callable set now always includes `request_tools`. Remote harness and bot clients get the same listing: a new `gbrain mcp provision` grant with the memory-reader, memory-writer or coding-agent profile and a new `gbrain agent register` daily-driver or coding-agent client keep the full callable surface but list the verbs and `request_tools` (`auth list-clients` shows `full (set by install)`); `gbrain auth rescope-client <id> --surface full` lists everything, and clients the operator pinned before keep their listing. Plugin manifests and a bare `serve` are unchanged.' },+ { since: '0.60.452.0', id: 'sources_add_federation_warning_gbra156', text: '`gbrain sources add` without `--federated` or `--no-federated` now warns on stderr that the new source is not federated, so a remote token or agent connection with no source grant cannot read its pages, and names `gbrain sources federate <id>` and `gbrain auth rescope --token <name> --sources <id>`. The stored default is still non-federated.' },+ // GBRA-187 harness-eval items (HARN-2, HARN-5). `since` is stamped at the merge slot.+ { since: '0.60.456.0', id: 'skillpack_skill_name', text: 'list_brain_skillpack now names each skill `name` instead of `slug`, and get_skill with source_id returns `name` too, matching the get_skill parameter it feeds. A skill name is not a page slug; clients that read `skills[].slug` read `skills[].name` instead.' },+ { since: '0.60.456.0', id: 'openclaw_core_scaffold', text: '`gbrain skillpack scaffold --harness openclaw` now installs five brain-first skills (query, brain-ops, capture, data-loss-gate, resolve-before-asking) instead of the full bundle, because OpenClaw lists every workspace skill in every prompt. `--all` installs the full bundle; existing workspaces are not changed.' },+ { since: '0.60.457.0', id: 'strict_landed_before_row_receipt', text: 'On a source configured `canonical: git`, a write whose door commit landed in Git before its database row and that then ends without committing (its grant revoked before the replay) now keeps `outcome.persistence = {git_state: committed, git_commit, landed_before_row: true}` on its receipt instead of an empty outcome, because the indexer will import those committed bytes. Other sources are unchanged.' },+ { since: '0.60.459.0', id: 'claim_refuses_broken_worktree_link', text: 'Claiming a directory as a canonical worktree (a source add or claim with a path) now refuses `git_worktree_link_broken` when the directory is a linked Git worktree whose checkout moved or whose repository entry was pruned; it used to claim it, and its door commits would have gone nowhere. Run git worktree repair from the repository, or add the worktree again, then rerun.' },+ // GBRA-190 G4 (fact fences). `since` is stamped at the merge slot.+ { since: '0.60.460.0', text: 'A legacy fact row whose `notability` or `confidence` is empty (a long-lived database whose column lost its NOT NULL) is now adopted into its page\'s `## Facts` table with the column default (notability `medium`, confidence 1) instead of being refused as unadoptable (`parse`) on every run, which held its page out of fact reconciliation and booked an `integrity_guard` halt each cycle. The fence cell is written first and the row derives its values from it. A row with no kind or visibility is still refused.' },+ { since: '0.60.460.0', text: '`gbrain repair duplicate-facts` keeps the copy that is a row of a page\'s `## Facts` table over an older database-only copy of the same fact, so the legacy copy is superseded onto the fence row instead of holding it (which left the legacy row unfenced and its page out of fact reconciliation). With no fenced copy it still keeps the oldest row. It stays preview-first and explicit.' },+ { since: '0.60.460.0', text: 'Dream consolidation on a managed brain now closes a fenced fact in its page: the closing `valid_until` is written into the fact\'s row in the entity page\'s `## Facts` fence, published with the take, and the database row takes its value from that cell, so a later sync or re-import keeps the fact closed. Before, only the database row was closed and the next re-import of the page reopened it. A fact with no fence row is still closed in the database. A fenced fact whose fence row is on another page is left open. Facts an earlier release closed in the database only are not changed (GBRA-190).' },+ // GBRA-207 (#7630). `since` is stamped at the merge slot.+ { since: '0.60.469.0', id: 'sync_refuses_relative_anchor', text: 'A sync whose source directory (`sources.local_path`) or legacy `sync.repo_path` is a relative path such as `.` now refuses `sync_anchor_relative` (exit 3) before any Git probe or import, instead of treating whatever folder it ran from as the brain repo (Git init, a baseline commit of the whole folder, an import of its markdown). `gbrain sources set-path <id> <absolute-directory>` sets the path, `gbrain sources set-path <id> --clear` now also clears a relative path on any source, and doctor\'s `default_source_local_path` warns on any source with a relative path, and `gbrain config set sync.repo_path` refuses one. Until the path is set, page and fact writes skip the file (database only), and so do page writes of any other source without its own path that write under `sync.repo_path` (that relative path could name this tree), matching a folder to its source ignores it, the dream, autopilot and maintenance cycles run only database phases (a cycle whose directory matches no source refuses its sync phase while `default`\'s path is relative), fence repair edits stored pages instead of files, `gbrain sources resolve-path` refuses it (exit 3), managed activation refuses it, `autopilot --install` needs `--repo`, a remote `submit_job` is refused and a managed brain records no managed root there. A few other local commands may still read it from the folder they run in, so set the path before relying on them.' },+ { since: '0.60.469.0', text: 'A foreground write (`remember`, `put_page`, `add_tag`, `takes_add` and the other caller-submitted page mutations) that ends with a retryable fault such as `preparation_stalled` now carries a diagnosis that inspects the writer read-only (`gbrain sources writer status --source <id> --json`) and says to resubmit the same write with a new request_id, instead of telling the user to run `gbrain sync`. A managed write (an import, repair, quarantine clear or grandfather) says to rerun its command, an operation such as `extract_facts` or writer administration gets the read-only inspect alone, and a canonical reconcile, a maintenance job (stored as `submit_job`) and a recorded sync fault keep their sync retry. `gbrain sources writer failures --group-by category` splits these by operation, and `--operation` narrows a listing to one; a row for a reconcile or managed `put_page` now reads `canonical_reconcile` or its `managed_*` intent in `operation`.' },+ // GBRA-95 RLS backfill. `since` is stamped at the merge slot.+ { since: '0.60.463.0', id: 'rls_backfill_unguarded_tables', text: 'On Postgres, a schema migration now enables row-level security on every gbrain table still without it. A table could be left without RLS because its RLS came only from a full schema replay, which `gbrain apply-migrations --yes` does not run (the four incremental-derivation tables, `page_segment_derivations`, `page_derivation_spend` and the two purge tables), or only from the auto-RLS event trigger, which needs a superuser and is absent on many managed brains. Supabase gives public tables anon and authenticated grants, so with the Data API on such a table was reachable with the anon key. Only a table still without RLS changes; one with a `GBRAIN:RLS_EXEMPT` comment is left alone; no row is read or written. Under a role that does not bypass RLS or does not own a table, the migration names the table on stderr and leaves it, and `gbrain doctor` (rls) keeps naming it with the fix. `gbrain repair missing-tables` now recreates a cache table with RLS on. A role that neither owns a table nor bypasses RLS reads no rows from it and cannot insert or update it once RLS is on (a read-only reporting role or a scoped write role, for example): give that role policies before upgrading (docs/ENGINES.md has the source-scope recipe), or, to keep a table open to it on purpose, add the `GBRAIN:RLS_EXEMPT reason=...` comment to that table. A write role set up with an earlier copy of that recipe also needs its new policies on `page_derivation_spend` and the two purge tables: without them its segmented derivation writes are refused, and the derivation purge triggers its writes fire read no purge, so they do not hold back a purged claim.' },+ { since: '0.60.464.0', text: 'On Postgres, `gbrain serve --access read-only` (stdio) no longer acts as a resident owner: it holds no session-lane (direct pool) connection, starts no persistence consumer, takes no on-commit notifier lease, runs no startup or idle sweep or facts drain, and refuses delegated sync and sweep requests (the CLI then runs them itself). Resolve, turn_context and context_pack still answer. A read-write serve, `gbrain jobs work` or the autopilot owns those duties; `gbrain doctor` reports when no process on the host holds the notifier (`commit_notifier`) or a consumer (`two_consumers_on_host`). Read-write and HTTP serves, and any serve on PGLite (the process that owns the database file), are unchanged.' },+ { since: '0.60.464.0', text: 'Security: `gbrain serve --access=read-only` (the `=` form) now sets the read-only ceiling. It passed flag validation before but started the serve with full write access, because only `--access read-only` was read. A config that used the `=` form now gets the read-only tool set and refused mutations it asked for; a repeated `--access` never widens a read-only value, and an invalid value in either form refuses. `--surface=<v>` likewise now sets the serve surface (it was ignored, leaving the full surface), and a repeated `--surface` takes the narrowest value.' },+ { since: '0.60.464.0', text: 'A managed sync pass whose process was killed or crashed now reads `resumable` / `pass_interrupted` in `gbrain sync status` (with an `interrupted` block naming the pass and the dead holder) instead of `pass_unfinished`, and a second killed pass reads `blocked` / `pass_abandoned` at once. `gbrain doctor` `managed_sync_blocked` lists both (warn, fail after an hour). A live pass still committing is no longer counted abandoned because its row is old.' },+ { since: '0.60.465.0', text: 'On a managed brain, `gbrain trust release h<id>` of a held fact whose entity page exists now writes the fact into that page\'s `## Facts` fence through the page-write door (a `trust_release` page write) and flips the hold in the same publication, so the released fact lives in the page\'s file and a re-import keeps it. Before, the release inserted a database-only row the file never held; a fence-guard hold that names an existing fence row created a database-only twin of it, and now indexes that row and leaves the file unchanged. The row is stamped `user_confirmed` with origin `trust_release` and the page keeps its trust tier. A hold decided after the release was accepted fails `preview_changed` and writes nothing; a remote request, or a deleted or private entity page, refuses and the fact stays held. A hold with no entity page, and unmanaged brains, release database-only as before.' },+ { since: '0.60.467.0', id: 'cli_flag_equals', text: 'A CLI flag value written with `=` that a command used to ignore now exits 2 (`invalid_params`) and names the two-argument spelling the command reads: `gbrain auth create --scopes=read` minted a full-access token. A switch never takes a value (`--yes=n`, `--dry-run=true`, `--apply=false` and `-y=n` exit 2 instead of being ignored). The `=` spellings a command reads on every path keep working, and so do the spending cap and plan hash the consent check of a command reads; a few that only some paths read now exit 2 with the two-argument fix, among them the source id given that way to `gbrain extract` for stale extraction, the checkout path given that way to the database-free `gbrain sources pull`, and the values given that way to the managed `gbrain sources` lifecycle subcommands (add, remove, archive, restore, purge, set-path, reclone). A value written with `=` on a flag that the help of the command declares used to be ignored and now applies, so check scripts that pass these that way: an all-interfaces bind address given to `gbrain serve` bound loopback and now listens on every interface; a public URL given to `gbrain serve` now sets the OAuth issuer and resource, so clients authorized under the old localhost issuer must reconnect, and admin cookies become Secure on https; the serve port, token lifetime and stdio idle timeout, the sync interval, source and repository, the import worker count, the autopilot install repository and supervisor, and the target engine and URL the legacy `gbrain migrate` copier reads take effect the same way; a spend cap given to `gbrain onboard` (off included) now applies, and one given to a resumed `gbrain doctor` remediation overrides the cap its checkpoint recorded; and `gbrain apply-migrations` now honors an accepted reviewed inventory and a forced retry of a wedged migration. An upgrade restart of a `gbrain jobs supervisor` started with values written that way (which it ignored) starts it again without them. A script that passes a value with `=` to a command that does not read that form now stops with the fix instead of running without the value.' },+ // Bun floor 1.4.4 (#7644, #7596). `since` is stamped at the merge slot.+ { since: '0.60.468.0', id: 'bun_floor_1_4_4', text: 'GBrain now requires Bun 1.4.4 or newer (previously 1.4.0). On Bun 1.4.0 to 1.4.3 every command except `gbrain --version` exits 1 before it starts with `GBrain requires Bun 1.4.4 or newer (found Bun <version>)` and the fix (`bun upgrade`, then restart GBrain; `gbrain post-upgrade` finishes an interrupted upgrade); `--version` prints the same refusal and exits 0. On Bun 1.4.0 to 1.4.2 a postgres.js connection over TLS keeps every chunk it receives, so a full sync or conversion pass over a TLS Postgres or pooler grew by about the bytes it read until memory guards killed it (#7644, #7596); 1.4.4 also carries the Brotli fix for CVE-2026-111497. Standalone gbrain binaries already embed Bun 1.4.4; a source or `bun install` install runs `bun upgrade` first, and `gbrain upgrade` on an older host Bun stops (autopilot holds) with the same fix.' },+ // #7632 function search_path. `since` is stamped at the merge slot.+ { since: '0.60.470.0', id: 'function_search_path_pg_temp', text: 'Every gbrain function that pins a search_path now ends it with `pg_temp` (`ALTER FUNCTION ... SET search_path`, no body changes). Before, the purge and withdrawal guards pinned `pg_catalog, public`, which leaves the session\'s temporary schema first for tables and views: any session, the owner included, could create an empty temp table or view named like a ledger (`fact_purges`, `page_purges`, `take_purges`, `fact_withdrawals`, `page_segment_derivation_purges`) and store a purged claim, page, take or derivation again, or save a withdrawn fact active. The guards are otherwise unchanged (SECURITY INVOKER, same bodies, same timing). Only gbrain\'s own functions change; another application\'s functions in the same database keep their search_path. Still open, not fixed here: on Postgres, a write role with no RLS policy on a ledger reads it empty, so its purged or withdrawn writes still go through (gbrain#7659). Where the migrating role does not own a function, the migration leaves it, names it on stderr and carries on, and `gbrain doctor` (`function_search_path`) prints the `ALTER FUNCTION` to run as its owner; that changes no owner, so later boots still succeed. An older gbrain binary that replays its schema against the brain (`gbrain init` or `gbrain upgrade` from it; an ordinary boot replays nothing) drops `pg_temp` from the functions it re-creates; doctor flags them, and `gbrain init` or `gbrain upgrade` with this release, or the printed `ALTER FUNCTION`, puts it back. An ordinary boot of this release does not.' },+ // #7705 (strict door planned state). `since` is stamped at the merge slot.+ { since: '0.60.471.0', id: 'door_planned_state', text: 'On a source with `canonical: git`, a page write now commits exactly the bytes it published, or nothing. A file edited on disk while the write was publishing refuses `git_target_changed`: nothing is committed and the edit stays on disk; when the file holds neither the write\'s bytes nor the version before it, the source\'s writes pause until `gbrain repair stuck-recovery` resolves it, and when it holds the version before the write, the write fails with nothing held (before, Git committed the edit while the database recorded the write\'s content). A staged change of the page\'s file refuses `git_index_conflict`, with the index untouched; commit or unstage it, then retry (before, the write replaced the staged change in the index). When the index still holds the version from before an earlier gbrain commit whose index update failed, the next write restores it and says so with a `git_index_restored` notice.' },+ // #7679 / #7685 orphaned worktree refresh. `since` is stamped at the merge slot.+ { since: '0.60.472.0', id: 'refresh_orphan_7679', text: 'A managed `gbrain sources refresh` whose process died (for example a container replaced mid-refresh) no longer wedges the checkout: the next plain `gbrain sources refresh` on the owner host finishes the orphaned refresh through the `--resume` path before its own refresh and reports it as `recovered`, and the owner completes an orphaned `syncing` refresh once every member has synced to its target. A refresh whose holder is still running, or is recorded on another host, is never taken over. `gbrain sources writer status` and `gbrain sync status` now name the holder and its `liveness` (`live`, `orphaned`, `unknown`) with a fix naming `--resume`. With `--json`, a refusal that carries a structured fix (`refresh_in_progress`, `refresh_diverged`, a git read failure) now prints `fix` as an object (argv, actor, next) instead of a string; the prose stays in `suggestion`, so read that when a script needs text. A plain refresh of a checkout whose interrupted refresh needs a reset decision now refuses `refresh_recovery_required` (it refused `refresh_in_progress`).' },+ // #7814 retired v0.13.1 grandfather. `since` is stamped at the merge slot.+ { since: '0.60.479.0', id: 'v0_13_1_grandfather_retired', text: 'Upgrading a brain whose migration ledger never recorded v0.13.1 (one created by `gbrain init` before v0.60.412.0, or opened from a new GBRAIN_HOME) no longer writes `validate: false` into every page\'s frontmatter: `gbrain apply-migrations` records v0.13.1 complete and changes no page, so no content_hash moves and a repo brain\'s files and Git history stay as the user wrote them. The stamp only opted a page out of the BrainWriter validators, which never refuse a write, and out of `gbrain integrity` scans, so on such a brain those scans now include the pages a fresh brain would scan. A page that already carries `validate` keeps it, and a grandfather write an older gbrain left queued ends `conflict` without writing. Pages an older release stamped keep the stamp; their pre-stamp frontmatter is in `~/.gbrain/migrations/v0_13_1-rollback.jsonl` on the machine that ran it.' },+ // #7617 / #7618 / #7644 / #7655 conversion. `since` is stamped at the merge slot.+ { since: '0.60.480.0', id: 'git_conversion_withhold_progress_push', text: '`gbrain sources canonical <source> --to git --apply` now converts a source with private pages instead of refusing: a private page whose committed file differs from it has the file\'s removal committed and stays in the database (`withhold` in the preview; before, the apply refused `git_conversion_blocked`), a private page whose file was removed by hand is withheld instead of deleted by the strict pass, and a private page whose committed file still equals it keeps that file in Git, as before. The apply prints a progress line on stderr at each phase and every 10 seconds, and prints a partial report on stdout when stopped by SIGTERM, SIGINT or SIGHUP (exit 143, 130 or 129). The source is git-canonical once `recordConversion` runs before strict_pass_2; the runtime report is authoritative: a stop before that leaves the source mixed, and a new preview lists what is left. Where the source pushes its commits, the apply pushes its own; a push that does not land is named in the receipt with the command that pushes now. One path Git cannot check no longer discards every adoption candidate, and holds clear only after the strict recheck passes.' },+ { since: '0.60.481.0', id: 'synthesize_session_source_7805', text: 'Dream synthesis writes a captured session only into the source its session recorded (the `.src-<id>` name in the corpus `sourced/` folder, or the session\'s frozen record), never into the source the cycle runs for. A session of another source waits for `gbrain dream --source <id> --phase synthesize`; a session whose source never resolved, and a file written before sessions recorded their source on a brain with more than one source, is held, already-extracted ones included, until `gbrain sweep --assign-corpus <source>` previews and `--apply` maps it. The phase summary and `details.source_routing` count both. Pages an earlier release already wrote into the wrong source are not moved (#7805, reported by @exGeni).' },+ { since: '0.60.481.0', id: 'put_visibility_not_applied_7841', text: '`put_page`, `edit_page` and `capture` content that asks for `visibility: private` in a form gbrain does not apply (`Private`, `PRIVATE`, or a `Visibility` key, however YAML spells it) is refused with `visibility_not_applied` instead of committing a world-readable page. Write `visibility: private` on its own frontmatter line, lowercase. Nothing ever widens a page\'s visibility (#7841).' },+ // #7846 extract_atoms session source. `since` is stamped at the merge slot.+ { since: '0.60.482.0', id: 'extract_atoms_session_source_7846', text: 'Dream `extract_atoms` writes a captured session\'s atoms only into the source its session recorded (the `.src-<id>` name in the corpus `sourced/` folder, or the session\'s frozen record), never into `default`. Each source\'s run now reads the session corpus and keeps only its own sessions (`gbrain dream --source <id> --phase extract_atoms`); meeting transcripts stay with `default`. A session whose source never resolved, and a file written before sessions recorded their source on a brain with more than one source (already-swept ones included), is held until `gbrain sweep --assign-corpus <source>` previews and `--apply` maps it. The phase summary and `details.source_routing` count both. Atoms an earlier release already wrote into the wrong source are not moved (#7846).' },+ // #7868, #7850, #7869, #7870, #7871 frozen-source privacy wave. `since` is stamped at the merge slot.+ { since: '0.60.483.0', id: 'corpus_session_tier4_only_7868', text: 'A captured session whose source was never recorded is no longer filed under `sources.default`, the only non-default source or `default` when the brain resolves it later: only a registered source whose folder contains the session\'s working directory counts. Any other session stays held as `corpus_source_unresolved` until `gbrain sweep --assign-corpus <source>` previews and `--apply` maps it. A brain with one source still uses that source, without recording it, so a source added later holds the session instead. Sessions an earlier release already recorded keep their source (#7868).' },+ { since: '0.60.483.0', id: 'synthesize_session_recheck_7850', text: 'A dream synthesis job checks its session\'s source again when it runs and is refused (`session_source_changed`) before any model call when the session now belongs to another source or its source was deleted and recreated; on a managed brain, jobs queued by an earlier release are checked too. `gbrain dream --dry-run` no longer records a session\'s source while previewing synthesize or extract_atoms (#7850).' },+ { since: '0.60.483.0', id: 'atom_retry_resume_session_source_7869_7870', text: 'An explicit atom retry (`gbrain jobs submit extract-atoms-drain` with `retryRequestId`) of a captured session that does not belong to the retry\'s source is refused with `transcript_source_mismatch` and starts nothing. A crashed atom generation whose pending members came from such a session is not resumed: the members are held, their atom pages left as they are, and counted in `details.source_routing.generation_members_held` (#7869, #7870).' },+ { since: '0.60.483.0', id: 'corpus_realpath_7871', text: 'Dream synthesis and extract_atoms recognize a session-corpus file through symlinks (a meeting folder or corpus path that points into the corpus, a linked `sourced/` folder), so such a file is routed to its session\'s source instead of being read as an ordinary transcript; a listed file whose real path cannot be read is held as `corpus_path_unresolvable` (#7871).' },+ // #7848 non-canonical private requests on file ingestion. `since` is stamped at the merge slot.+ { since: '0.60.483.0', id: 'visibility_noncanonical_hold_7848', text: 'Git sync (classic and managed) and `gbrain import` (with or without a Git checkout) hold a file whose frontmatter asks for `visibility: private` in a spelling page privacy does not read (`Visibility: private`, `visibility: Private`, a block scalar, a list, a flow mapping or a merge key) as `visibility_noncanonical` instead of storing a world-readable page; the rest of the source syncs and an existing page keeps its last imported version. Write `visibility: private` on its own frontmatter line (lowercase) and commit, or import the file again. Doctor `git_held_files` counts the held files and the stored pages that already ask for private this way (`details.visibility_noncanonical`); nothing rewrites them. On a brain without managed persistence, `gbrain put --file` and `gbrain capture --file` of a page over the single-string ceiling refuse such a request with `visibility_not_applied`, as a smaller write does.' },+ // #7880 ingest_capture and company-brain imports on a brain without managed persistence. `since` is stamped at the merge slot.+ { since: '0.60.485.0', id: 'ingest_capture_private_hold_7880', text: 'On a brain without managed persistence, an `ingest_capture` job (the webhook `POST /ingest` route and the ingestion daemon) whose content asks for `visibility: private` in a spelling page privacy does not read (`Visibility: private`, `visibility: Private`, a block scalar, a list, a flow mapping or a merge key) is refused `visibility_not_applied` and dead-letters on its first attempt instead of storing a world-readable page, as a managed brain already refused it. A legacy company-brain connect or sync holds such a file as `visibility_noncanonical` instead of importing it world-readable, and the GitHub and Google connectors hold it on a database-only source as they do with a checkout. Write `visibility: private` on its own frontmatter line (lowercase) and send it again. Nothing rewrites pages already stored (#7880).' },+ // #7897 the remaining content imports without the private-request hold. `since` is stamped at the merge slot.+ { since: '0.60.487.0', id: 'private_hold_sweep_7897', text: 'Three more writers refuse content that asks for `visibility: private` in a spelling page privacy does not read (`Visibility: private`, `visibility: Private`, a block scalar, a list, a flow mapping or a merge key) instead of storing a world-readable page. `gbrain repair url-clone-fences` lists such a clone file under `visibility_noncanonical`; apply keeps its fence rows database-only and restores the file from HEAD so the clone can pull again, but does not import HEAD as the page and reports `held`, not `restored` (an approval previewed before this release applies the same way). On a brain without managed persistence, `gbrain brainstorm` / `gbrain lsd` `--save` refuses it `visibility_not_applied` and exits 1, as a managed brain already did, and a save whose import is refused for another reason is reported not saved. `gbrain sources reconcile` refuses a merge that takes a file\'s `Visibility: private` with `visibility_not_applied`. Write `visibility: private` on its own frontmatter line (lowercase) and try again. Nothing rewrites pages already stored (#7897).' },+ // #7900 v0.53.0 content checkpoint after a host move. `since` is stamped at the merge slot.+ { since: '0.60.488.0', id: 'content_checkpoint_host_move_7900', text: 'After a brain moves hosts, `gbrain apply-migrations` at v0.53.0 (`content-checkpoints`) still refuses `The canonical root changed after setup began` when an unfinished shared-skills content setup checkpoint names the old host\'s root, but its fix is now `gbrain repair content-receipt-root --source <source>` (before, it told you to resume at the old root). That repair, and doctor `sync_freshness` (`content_receipt_stale`), now list unfinished checkpoints that name another root as well as complete receipts (before, the repair found nothing). The preview names the step that proves the new root; for a source claimed on the new host that is a self-transfer on the owner host, which rebinds the checkpoint, followed by `gbrain sources writer activate --confirm-quiesced --shared-skills` when shared skills are on. Then `gbrain apply-migrations --force-retry 0.53.0` completes. Nothing is rewritten without those explicit steps.' },+ // GBRA-205: Garry approved the default ("Yes", 2026-10-26), on the sealed held-out result (gbrain-evals docs/benchmarks/2026-10-26-gbrain-8k-conversion.md).+ { since: '0.60.491.0', text: '`query` with a token budget now puts question-ranked dated saved facts ahead of the pages (the facts block): up to 22% of the budget and 40 rows, oldest first, each with its date header, in place of the facts arm (and of the temporal fact reserve when it is on). On a sealed held-out set of 200 questions it answered 183 against 151 without it. It adds about 70 tokens to a median budgeted result (p95 about 700) and no model call. Queries without a budget, `search` and `think` are unchanged; remote callers see only facts whose source pages they could read. `gbrain config set search.query_facts_block false` restores the previous results.' },+ // GBRA-205 saved-facts wave (harness case cr-05). `since` is stamped at the merge slot.+ { since: '0.60.493.0', id: 'saved_facts_stems', text: 'Search `saved_facts` matches a saved fact\'s words the way the facts keyword index does: whole words after stemming with the brain\'s full-text language, so a query for "seats renewal" finds a fact that says "renews at 52 seats", and "count" no longer matches inside "countersigned". A fact still needs three quarters of the query\'s words, and never more than four, so a question asked in full finds the short fact that answers it, not counting words that frame a question about a figure (count, number, total, amount); words of a CJK script still match as substrings. Among facts matching as many words, the newest comes first, and at most five facts ride a reply. On by default (`gbrain config set search.saved_facts_supersede false` turns it off): search also returns a newer fact with a correction cue that shares two matched words with a returned fact, ahead of the fact it corrects, which is labelled `corrected_by`; every such correction comes back, not only the newest, and an acknowledgement that adds no figure (dates left out) corrects nothing, and a reply that only reports a correction ("corrected", "correcting") makes none; nor does a cue fact saved under another entity, of lower trust, or sharing none of the names the older fact matched.' },+ // GBRA-188. since: '0.60.495.0' is stamped at the merge slot.+ { since: '0.60.495.0', id: 'brain_first_instructions_gbra188', text: 'Harness installs now write a short marked brain-first block into the harness\'s own instruction file by default (Claude Code ~/.claude/CLAUDE.md, Codex AGENTS.md or a non-empty AGENTS.override.md, opencode AGENTS.md, the OpenClaw workspace AGENTS.md, an existing Hermes SOUL.md, a Cursor project rule): `gbrain setup <harness>`, `gbrain connect --install`, `gbrain bootstrap harness` and the thin-client GBRAIN-INSTRUCTIONS.md. Text outside the markers is never touched and a re-run upgrades the block in place. `--no-instructions` or GBRAIN_AGENT_INSTRUCTIONS=0 removes it and is remembered; `--remove` strips it; `gbrain setup <harness> --instructions-only` manages only the block. doctor `harness_wiring` warns when a registered harness\'s block is missing, stale or damaged.' },+ { since: '0.60.495.0', id: 'install_surface_starter_gbra188', text: 'New stdio registrations gbrain writes (`gbrain setup`, `gbrain bootstrap hooks`, the readiness and `gbrain init` registration lines) now pin `--surface starter`, the daily-driver tool list, instead of `--surface full` (Garry, 2026-10-24; gbrain-evals#191: reach and accuracy unchanged, 17-23% fewer input tokens). An existing registration keeps its surface and is never rewritten; the Codex and Claude Code plugins and a bare `gbrain serve` stay on full. `--surface full` is the opt-in for every operation, any parameter the starter list leaves out is still accepted, and `request_tools` returns full schemas or widens the session to a left-out tool. doctor `harness_wiring` notes a full registration as information only.' },+ { since: '0.60.495.0', id: 'brain_first_turn_rule_gbra188', text: 'The brain-first rule now rides every turn, not only the system prompt (Garry, 2026-10-25). The block (v2) opens with "before answering any turn, check GBrain first"; while it is installed, the Claude Code and Codex UserPromptSubmit hook leads each turn with that one-line rule, with or without a context block; `gbrain setup openclaw` writes a hook-only plugin (gbrain-turn-rule, before_prompt_build prependContext) under the OpenClaw state dir and installs it with the openclaw CLI (install --link --accept-capabilities, enable, plugins.entries.gbrain-turn-rule.hooks.allowConversationAccess true), or prints those commands with no openclaw on PATH; and `gbrain setup hermes` adds a marked, self-contained `pre_llm_call` shell hook (`/bin/echo` of the rule) to $HERMES_HOME/config.yaml and approves that command in Hermes\'s shell-hooks-allowlist.json, or prints the entry when config.yaml already has a `hooks:` section. `--no-instructions` or GBRAIN_AGENT_INSTRUCTIONS=0 turns the rule off with the block.' },+ // #7891 provenance-stamp drift check and repair. `since` is stamped at the merge slot.+ { since: '0.60.496.0', id: 'provenance_stamp_drift_7891', text: 'New doctor check `provenance_stamp_drift` warns when a page\'s provenance stamps (`source_kind`, `ingested_via`, `ingested_at`) or its lowering `trust_tier` marker were removed or rewritten by a write that is not the owner\'s own (before v0.60.486.0, #7892, made the stamps server-owned), or when a classified page\'s file would re-read above its stored tier. It reads page history and write receipts only and writes nothing. New explicit-only repair `gbrain repair provenance --source <id>` previews each page with its stamps, marker and tier before and after; `--apply --expect <hash> --yes` puts the stamps back, stamps the lower marker, lowers the page to it and files a `lower_page` proposal that `gbrain trust review` confirms or reverts. It only lowers trust, never writes the owner\'s own edits, writes unattributed ones only with `--include-ambiguous`, and never runs from `--all`, the remediation plan or an upgrade.' },+ // #7917 / #7918 maintenance lowerings and unreadable derivation declarations. `since` is stamped at the merge slot.+ { since: '0.60.496.0', id: 'maintenance_lowering_7917_7918', text: 'A maintenance publication that lowers a page\'s trust tier (a dream or patterns write over a page the owner wrote, such as a grounding repair outside the output prefix) now files a `lower_page` proposal, so the page shows up in `gbrain trust review`; before, only agent content writes did. A maintenance write whose derivation declaration is present but unreadable now publishes at `unknown` and stamps an `unknown` trust marker into the file, instead of taking the write channel\'s tier and leaving the file unmarked. Both only lower trust. `gbrain trust backfill --repair-unknown-markers` and `gbrain repair provenance` on a classic (unmanaged) checkout now also write the repaired page to its file in the same step, and leave a page whose file holds an edit sync has not imported untouched (run `gbrain sync` first).' },+ { since: '0.60.498.0', id: 'plugin_turn_hook_gbra188', text: 'The Claude Code and Codex plugins (gbrain, gbrain-coding, gbrain-daily) now declare the UserPromptSubmit hook `gbrain setup` writes, run through the plugin launcher with GBRAIN_HOOK_LANE=plugin (Claude Code `hook user-prompt`, 3 s; Codex `hook user-prompt --harness codex`, 15 s, skipped until trusted in /hooks). While a brain-first block is installed it adds the per-turn rule, and with a serve running the read-context block; it exits without output when setup, bootstrap or a workspace install already wires the event, so a turn gets it once; a recorded `gbrain setup <harness> --no-hooks` turns it off too. doctor `harness_wiring` and `gbrain post-upgrade` now name `gbrain setup <harness> --instructions-only` for a plugin-only install with no brain-first block.' },+ { since: '0.60.502.0', id: 'provider_key_rules_7959', text: 'The secret scanner now catches OpenRouter (`sk-or-v1-`), Groq (`gsk_`), xAI (`xai-`), Perplexity (`pplx-`) and Replicate (`r8_`) API keys: `gbrain sources push` refuses a page that holds one, and search, recall and transcript import redact it. Rotate such a key with its provider rather than allowlisting it. `gbrain doctor` asks once for a fresh `gbrain transcripts audit-secrets` run.' },+ { since: '0.60.504.0', text: '`gbrain sync status` and doctor `managed_sync_blocked` no longer read a managed sync pass `pass_abandoned` while a process still renews its drain lease (a pass started over earlier interrupted ones), nor `pass_interrupted` when the pass ended between their pass and lease reads; liveness reads the database clock, and doctor keeps status\'s quiet-window rule when the lock read fails. A pass stopped by a lost database connection, a statement timeout, write back-pressure or a lost drain lease now ends `resumable` (`stop_reason` `transient` or `lease_lost`) instead of reading killed. The on-commit notifier, the notifier block and `sources status` `writer_heartbeat` no longer count a killed holder\'s lock row as a live holder, so the notifier starts the sync that takes its expired lease over.' },+ { since: '0.60.504.0', text: 'On Postgres, `gbrain serve --access read-only` (stdio) binds its own read socket for resolve, turn_context and context_pack instead of the source and legacy sockets, so a read-write serve started later for the same source always answers delegated sync, nudge and sweep requests; before, a read-only serve that bound first made the CLI fall back with `unsupported_kind`.' },+ { since: '0.60.505.0', id: 'turn_hook_every_install_gbra243', text: 'Every harness install path now wires the per-turn brain-first hook: `gbrain setup claude-code|codex --instructions-only` (after a plain `mcp add`), `gbrain connect --install`, `gbrain connect --harness … --install` and `gbrain bootstrap harness` (Codex, and Claude Code when its own hooks are off) add one UserPromptSubmit entry that prints only the "check gbrain first" rule and reads no brain data; it yields to setup, bootstrap and plugin hooks, so a turn still gets the rule once. Each install runs the hook (or `openclaw plugins inspect gbrain-turn-rule --runtime --json`, `hermes hooks doctor`) and exits non-zero with `turn_hook_not_loaded` when it does not load; doctor `harness_wiring` warns the same way. `--no-instructions` or `--no-hooks` installs none, and registrar mode stays MCP-only. On OpenClaw and Hermes the per-turn rule now names the gbrain tools and the `tool_search` route.' },+ // GBRA-156 harness wave (#7598). since is stamped at the merge slot.+ { since: '0.60.506.0', id: 'remote_page_cap_gbra156', text: 'A remote `search` or `query` that names no `limit` now returns every page that matched its words, a name or an exact identity, plus pages that matched only by meaning up to 3 pages in total (`search.remote_page_cap`, default 3; `0` turns it off). A line after the results says how many pages were left out, and `_meta.retrieval.page_cap` counts them. An explicit `limit`, a paged call, an image `query`, a host-run subagent tool call and every CLI call on the brain host (a thin-client CLI call is remote and gets the cap) return the same rows as before.' },+ { since: '0.60.506.0', id: 'sources_not_readable_refusal_gbra156', text: 'An unscoped remote `search`, `query`, `recall` or `list_pages` from a connection with no source grant that comes back empty, when every source it reads holds no page (for `recall` and `query`, no world fact either) and other sources do, is now refused with `sources_not_readable` (host fix `gbrain sources federate <source>` with the source the user picks, or a grant for the token) instead of answering empty with a degraded notice. A scope that holds pages but matched nothing, a brain with no pages, granted tokens, OAuth clients, bound connections and reads that name a source behave as before.' },+ { since: '0.60.506.0', id: 'sources_add_next_gbra156', text: '`gbrain sources add` without `--federated` or `--no-federated`, on a brain that is not managed where connections with no source grant read no page, now prints `next: gbrain sources federate <id>` on stdout. `--json` prints `{id, name, local_path, remote_url, federated, next?}` on stdout and moves the human lines to stderr. A managed brain prints its lifecycle receipt as before.' },+ // GBRA-156 #7601 quoted transcript rows. since is stamped at the merge slot.+ { since: '0.60.507.0', id: 'search_quotes_transcript_rows', text: '`search`, `query` and `assemble_evidence` now return a row from a conversation page (type `conversation`, what transcript imports write) with `quoted: "transcript"` and its text wrapped as a quoted excerpt: a line naming the page as source material, then a code fence. Tool-call and function-call JSON inside (Gemini `functionCall` included), and XML-style `function_calls`, `invoke`, `tool_use` and `tool_call` blocks, become `[tool call: <name>, arguments omitted]`. Under `return_unit: auto` an explicit `token_budget` pays for the fence, so a conversation-heavy result can return fewer or shorter rows. Stored pages, ranking, scores and other page types are unchanged; read `quoted` to tell these rows apart, and `get_page` for the text as stored.' },];/** The newest disclosed change's release: the notice id moves only when a release adds rows. */diff --cc src/core/cli-flag-registry.generated.tsindex f0f6d218ec,f71f7354e7..08048c2e3c--- a/src/core/cli-flag-registry.generated.ts+++ b/src/core/cli-flag-registry.generated.ts@@@ -94,9 -94,9 +94,9 @@@ export const CLI_FLAG_REGISTRY: Record<'orphans': ['--accept-pending', '--all', '--background', '--brain', '--count', '--cursor', '--explain', '--follow', '--force', '--help', '--include-pseudo', '--json', '--limit', '--mode', '--no-accept-pending', '--progress-interval', '--progress-json', '--quiet', '--source', '--timeout', '--timings', '--wait'],'pages': ['--brain', '--dry-run', '--force', '--help', '--json', '--limit', '--literal-pathspecs', '--name-only', '--older-than', '--quiet', '--retry-parked', '--show-toplevel', '--source', '--source-id', '--state', '--verify', '--yes'],'pglite-repair': ['--brain', '--break-lock', '--dry-run', '--expect', '--help', '--http', '--json', '--legacy-lock', '--path', '--probe', '--source', '--vectors', '--yes'],- 'post-upgrade': ['--accept-reviewed-inventory', '--acknowledge-no-backup', '--admin-intent', '--all', '--all-applied', '--apply', '--apply-clean-hunks', '--backup-confirmed', '--brain', '--catch-up', '--changed', '--channel', '--check', '--concurrency', '--confirm-quiesced', '--confirm-relocated-root', '--content-root', '--cost-estimate', '--embedding-dimensions', '--embedding-model', '--exit-code', '--expand', '--expected-epoch', '--expected-state', '--explain', '--export-db-only', '--export-source', '--ff-only', '--force', '--force-all', '--force-orchestrator', '--force-retry', '--force-schema', '--format', '--held', '--help', '--host-dir', '--images', '--inject-bootstrap', '--inline', '--install', '--interval', '--is-ancestor', '--json', '--limit', '--list', '--manifest', '--markdown', '--max-rss', '--max-usd', '--migrate-only', '--migration', '--mode', '--multimodal', '--name-only', '--no-autopilot-install', '--no-bun-floor-check', '--no-embed', '--no-expand', '--no-inject', '--no-restart', '--no-worker', '--non-interactive', '--now', '--once', '--only', '--oversized', '--path', '--pause', '--phase', '--quiet', '--reason', '--refs', '--remediation-plan', '--repo', '--require-db', '--restart-only', '--resume', '--retry-global', '--search-mode', '--self-transfer', '--skills-dir', '--skip-verify', '--source', '--stale', '--statement-timeout', '--status', '--stop-mcp', '--swap-only', '--tags', '--target', '--uninstall', '--user', '--vectors', '--verify', '--version', '--workers', '--yes'],- 'post-upgrade': ['--accept-reviewed-inventory', '--acknowledge-no-backup', '--admin-intent', '--all', '--all-applied', '--apply', '--apply-clean-hunks', '--backup-confirmed', '--brain', '--catch-up', '--changed', '--channel', '--check', '--concurrency', '--confirm-quiesced', '--confirm-relocated-root', '--content-root', '--cost-estimate', '--embedding-dimensions', '--embedding-model', '--exit-code', '--expand', '--expected-epoch', '--expected-state', '--explain', '--export-db-only', '--export-source', '--ff-only', '--force', '--force-all', '--force-orchestrator', '--force-retry', '--force-schema', '--format', '--held', '--help', '--host-dir', '--images', '--inject-bootstrap', '--inline', '--install', '--interval', '--is-ancestor', '--json', '--limit', '--list', '--manifest', '--markdown', '--max-rss', '--max-usd', '--migrate-only', '--migration', '--mode', '--multimodal', '--name-only', '--no-autopilot-install', '--no-bun-floor-check', '--no-embed', '--no-expand', '--no-inject', '--no-restart', '--no-worker', '--non-interactive', '--now', '--once', '--only', '--oversized', '--path', '--pause', '--phase', '--quiet', '--reason', '--refs', '--remediation-plan', '--repo', '--require-db', '--restart-only', '--resume', '--retry-global', '--search-mode', '--segmented', '--self-transfer', '--skills-dir', '--skip-verify', '--slug', '--source', '--stale', '--statement-timeout', '--status', '--stop-mcp', '--swap-only', '--tags', '--target', '--uninstall', '--user', '--value', '--vectors', '--verify', '--version', '--workers', '--yes'],++ 'post-upgrade': ['--accept-reviewed-inventory', '--acknowledge-no-backup', '--admin-intent', '--all', '--all-applied', '--apply', '--apply-clean-hunks', '--backup-confirmed', '--brain', '--catch-up', '--changed', '--channel', '--check', '--concurrency', '--confirm-quiesced', '--confirm-relocated-root', '--content-root', '--cost-estimate', '--embedding-dimensions', '--embedding-model', '--exit-code', '--expand', '--expected-epoch', '--expected-state', '--explain', '--export-db-only', '--export-source', '--ff-only', '--force', '--force-all', '--force-orchestrator', '--force-retry', '--force-schema', '--format', '--held', '--help', '--host-dir', '--images', '--inject-bootstrap', '--inline', '--install', '--interval', '--is-ancestor', '--json', '--limit', '--list', '--manifest', '--markdown', '--max-rss', '--max-usd', '--migrate-only', '--migration', '--mode', '--multimodal', '--name-only', '--no-autopilot-install', '--no-bun-floor-check', '--no-embed', '--no-expand', '--no-inject', '--no-restart', '--no-worker', '--non-interactive', '--now', '--once', '--only', '--oversized', '--path', '--pause', '--phase', '--quiet', '--reason', '--refs', '--remediation-plan', '--repo', '--require-db', '--restart-only', '--resume', '--retry-global', '--search-mode', '--segmented', '--self-transfer', '--skills-dir', '--skip-verify', '--slug', '--source', '--stale', '--statement-timeout', '--status', '--stop-mcp', '--swap-only', '--tags', '--target', '--uninstall', '--user', '--vectors', '--verify', '--version', '--workers', '--yes'],'pricing': ['--brain', '--help', '--input', '--json', '--max-cost', '--max-runtime', '--output', '--rate', '--source'],- 'projections': ['--brain', '--dry-run', '--enable-dcr', '--help', '--http', '--images', '--install', '--json', '--limit', '--now', '--only', '--port', '--public-url', '--remove', '--source', '--stale', '--status', '--surface', '--user'],+ 'projections': ['--advertise', '--brain', '--dry-run', '--enable-dcr', '--help', '--http', '--images', '--install', '--json', '--limit', '--now', '--only', '--port', '--public-url', '--remove', '--source', '--stale', '--status', '--surface', '--user'],'protocol': ['--access', '--add', '--all', '--allowed-operations', '--asof', '--body', '--bound-slug-prefixes', '--brain', '--client', '--cursor', '--date', '--dated-since', '--days', '--entities', '--federated-read', '--from-index', '--full', '--fuzzy', '--group-by', '--help', '--http', '--id', '--include-deleted', '--json', '--kind', '--limit', '--max-usd', '--model', '--multimodal', '--no-autopilot-install', '--page', '--params', '--path', '--phase', '--prefer', '--provenance', '--purge', '--recent', '--refresh-operations', '--row', '--scopes', '--session', '--since', '--slugs', '--source', '--sources', '--stale', '--surface', '--synthesize', '--target', '--to', '--token', '--updated-after', '--url', '--yes'],'providers': ['--brain', '--embedding-dimensions', '--embedding-model', '--help', '--json', '--model', '--source', '--status', '--touchpoint'],'publish': ['--accent', '--bg', '--border', '--brain', '--card-bg', '--code-bg', '--error', '--fg', '--help', '--json', '--link', '--muted', '--out', '--password', '--source', '--title'],@@@ -135,8 -135,8 +135,8 @@@'takes': ['--about', '--accept', '--applied', '--apply', '--batch-size', '--before', '--brain', '--bucket-size', '--by', '--claim', '--date', '--dim', '--dir', '--domain', '--dry-run', '--embedding-dimensions', '--embedding-model', '--evidence', '--expected-revision', '--expired', '--failed', '--force', '--from-pages', '--group-by', '--help', '--holder', '--include-covered', '--json', '--kind', '--limit', '--max-pages', '--min-confidence', '--min-n', '--no-embed', '--no-embedding', '--offset', '--outcome', '--path', '--quality', '--reject', '--request-id', '--retry-failed', '--row', '--semantic', '--since', '--slug', '--sort', '--source', '--source-id', '--stale', '--state', '--status', '--supabase', '--to', '--unapplied', '--unit', '--until', '--value', '--verdict', '--weight', '--who', '--yes'],'think': ['--anchor', '--brain', '--calibration-holder', '--help', '--json', '--mcp-only', '--model', '--only', '--reference-date', '--request-id', '--rounds', '--save', '--since', '--source', '--source-id', '--take', '--until', '--with-calibration'],'transcripts': ['--accept-pending', '--all', '--all-discovery', '--apply', '--background', '--brain', '--days', '--dry-run', '--embed', '--explain', '--facts', '--follow', '--format', '--full', '--help', '--include-self', '--json', '--limit', '--max-bytes', '--max-cost-usd', '--no-accept-pending', '--only', '--progress-interval', '--progress-json', '--purge', '--quiet', '--since', '--source', '--source-id', '--timeout', '--timings', '--wait'],- 'trust': ['--accept-all', '--accept-pending', '--all', '--background', '--batch-size', '--brain', '--break-lock', '--db-url', '--dry-run', '--expect', '--explain', '--follow', '--from', '--help', '--http', '--json', '--kind', '--legacy-lock', '--no-accept-pending', '--only', '--path', '--progress-interval', '--progress-json', '--quiet', '--reason', '--reason-family', '--relower', '--remove', '--repair-unknown-markers', '--resume', '--since', '--slug', '--source', '--source-id', '--timeout', '--timings', '--undo', '--uri-prefix', '--version', '--wait', '--yes'],- 'upgrade': ['--accept-reviewed-inventory', '--acknowledge-no-backup', '--admin-intent', '--all', '--all-applied', '--apply', '--apply-clean-hunks', '--backup-confirmed', '--brain', '--catch-up', '--changed', '--channel', '--check', '--concurrency', '--confirm-quiesced', '--confirm-relocated-root', '--content-root', '--cost-estimate', '--dry-run', '--embedding-dimensions', '--embedding-model', '--exit-code', '--expand', '--expected-epoch', '--expected-state', '--explain', '--export-db-only', '--export-source', '--ff-only', '--force', '--force-all', '--force-orchestrator', '--force-retry', '--force-schema', '--format', '--held', '--help', '--host-dir', '--images', '--inject-bootstrap', '--inline', '--install', '--interval', '--is-ancestor', '--json', '--limit', '--list', '--manifest', '--markdown', '--max-rss', '--max-usd', '--migrate-only', '--migration', '--mode', '--multimodal', '--name-only', '--no-autopilot-install', '--no-bun-floor-check', '--no-embed', '--no-expand', '--no-inject', '--no-restart', '--no-worker', '--non-interactive', '--now', '--once', '--only', '--oversized', '--path', '--pause', '--phase', '--quiet', '--reason', '--refs', '--remediation-plan', '--repo', '--require-db', '--restart-only', '--resume', '--retry-global', '--search-mode', '--self-transfer', '--skills-dir', '--skip-verify', '--source', '--stale', '--statement-timeout', '--status', '--stop-mcp', '--swap-only', '--tags', '--target', '--uninstall', '--user', '--vectors', '--verify', '--version', '--workers', '--yes'],+ 'trust': ['--accept-all', '--accept-pending', '--all', '--background', '--batch-size', '--brain', '--break-lock', '--db-url', '--dry-run', '--expect', '--explain', '--follow', '--from', '--help', '--http', '--json', '--kind', '--legacy-lock', '--no-accept-pending', '--only', '--path', '--probe', '--progress-interval', '--progress-json', '--quiet', '--reason', '--reason-family', '--relower', '--remove', '--repair-unknown-markers', '--resume', '--since', '--slug', '--source', '--source-id', '--timeout', '--timings', '--undo', '--uri-prefix', '--version', '--wait', '--yes'],- 'upgrade': ['--accept-reviewed-inventory', '--acknowledge-no-backup', '--admin-intent', '--all', '--all-applied', '--apply', '--apply-clean-hunks', '--backup-confirmed', '--brain', '--catch-up', '--changed', '--channel', '--check', '--concurrency', '--confirm-quiesced', '--confirm-relocated-root', '--content-root', '--cost-estimate', '--dry-run', '--embedding-dimensions', '--embedding-model', '--exit-code', '--expand', '--expected-epoch', '--expected-state', '--explain', '--export-db-only', '--export-source', '--ff-only', '--force', '--force-all', '--force-orchestrator', '--force-retry', '--force-schema', '--format', '--held', '--help', '--host-dir', '--images', '--inject-bootstrap', '--inline', '--install', '--interval', '--is-ancestor', '--json', '--limit', '--list', '--manifest', '--markdown', '--max-rss', '--max-usd', '--migrate-only', '--migration', '--mode', '--multimodal', '--name-only', '--no-autopilot-install', '--no-bun-floor-check', '--no-embed', '--no-expand', '--no-inject', '--no-restart', '--no-worker', '--non-interactive', '--now', '--once', '--only', '--oversized', '--path', '--pause', '--phase', '--quiet', '--reason', '--refs', '--remediation-plan', '--repo', '--require-db', '--restart-only', '--resume', '--retry-global', '--search-mode', '--segmented', '--self-transfer', '--skills-dir', '--skip-verify', '--slug', '--source', '--stale', '--statement-timeout', '--status', '--stop-mcp', '--swap-only', '--tags', '--target', '--uninstall', '--user', '--value', '--vectors', '--verify', '--version', '--workers', '--yes'],++ 'upgrade': ['--accept-reviewed-inventory', '--acknowledge-no-backup', '--admin-intent', '--all', '--all-applied', '--apply', '--apply-clean-hunks', '--backup-confirmed', '--brain', '--catch-up', '--changed', '--channel', '--check', '--concurrency', '--confirm-quiesced', '--confirm-relocated-root', '--content-root', '--cost-estimate', '--dry-run', '--embedding-dimensions', '--embedding-model', '--exit-code', '--expand', '--expected-epoch', '--expected-state', '--explain', '--export-db-only', '--export-source', '--ff-only', '--force', '--force-all', '--force-orchestrator', '--force-retry', '--force-schema', '--format', '--held', '--help', '--host-dir', '--images', '--inject-bootstrap', '--inline', '--install', '--interval', '--is-ancestor', '--json', '--limit', '--list', '--manifest', '--markdown', '--max-rss', '--max-usd', '--migrate-only', '--migration', '--mode', '--multimodal', '--name-only', '--no-autopilot-install', '--no-bun-floor-check', '--no-embed', '--no-expand', '--no-inject', '--no-restart', '--no-worker', '--non-interactive', '--now', '--once', '--only', '--oversized', '--path', '--pause', '--phase', '--quiet', '--reason', '--refs', '--remediation-plan', '--repo', '--require-db', '--restart-only', '--resume', '--retry-global', '--search-mode', '--segmented', '--self-transfer', '--skills-dir', '--skip-verify', '--slug', '--source', '--stale', '--statement-timeout', '--status', '--stop-mcp', '--swap-only', '--tags', '--target', '--uninstall', '--user', '--vectors', '--verify', '--version', '--workers', '--yes'],'waiting': ['--all', '--as-of', '--brain', '--dry-run', '--fail-fast', '--help', '--json', '--note', '--source', '--stale-ok', '--status', '--tools-json', '--top', '--type'],'watch': ['--brain', '--help', '--json', '--max-pages', '--min-confidence', '--source', '--window-turns'],'whoknows': ['--access', '--add', '--all', '--allowed-operations', '--asof', '--body', '--bound-slug-prefixes', '--brain', '--client', '--cursor', '--date', '--dated-since', '--days', '--entities', '--explain', '--federated-read', '--from-index', '--full', '--fuzzy', '--group-by', '--help', '--http', '--id', '--include-deleted', '--json', '--kind', '--limit', '--max-usd', '--mcp-only', '--model', '--multimodal', '--no-autopilot-install', '--page', '--params', '--path', '--phase', '--prefer', '--purge', '--recent', '--refresh-operations', '--request-id', '--row', '--scopes', '--session', '--since', '--slugs', '--source', '--sources', '--stale', '--surface', '--to', '--updated-after', '--url', '--yes'],