Skip to content

A self-improving RLM agent for coding workflows and long-running autonomous tasks.

Rust

release: v0.10.2 (#3789)

prime-agent-release-bot[bot]ea9d9602ec09
diff --git a/.changes/async-turn-contract.md b/.changes/async-turn-contract.md
deleted file mode 100644
index eef89feafd..0000000000
--- a/.changes/async-turn-contract.md
+++ /dev/null
@@ -1 +0,0 @@
-- The static usage layer no longer tells agents to await commands before ending the turn: since #3580 daemonless sessions also wake on a queued `[bash-done]` turn, every session kind has completion notifications, and the no-polling rule (end the turn and collect the result on the done notification) is the one turn-end contract.
diff --git a/.changes/azure-responses-v1-route.md b/.changes/azure-responses-v1-route.md
deleted file mode 100644
index 4df89bbd79..0000000000
--- a/.changes/azure-responses-v1-route.md
+++ /dev/null
@@ -1 +0,0 @@
-- Azure OpenAI Responses requests no longer fail with 404 on the v1 API: they now post to `<resource>/openai/v1/responses?api-version=v1` with the deployment as the request's `model`, instead of a `/deployments/<name>/responses` route Azure does not serve. A different `AZURE_OPENAI_API_VERSION` (for example `preview`, or a dated version against a custom `/openai` base URL) is sent as the `api-version` query instead, and an empty one falls back to `v1`.
diff --git a/.changes/entry-inputs.md b/.changes/entry-inputs.md
deleted file mode 100644
index 7cfc5c8023..0000000000
--- a/.changes/entry-inputs.md
+++ /dev/null
@@ -1 +0,0 @@
-- Factory machines can loop back into an entry state: an entry state may now declare inputs as long as all of them are optional. Its first entry binds the null sentinel, and a later entry through a transition binds the real output, so an investigate/verify loop no longer needs a duplicate re-investigate state. A required input on an entry state is still rejected, with a message saying it must be optional.
diff --git a/.changes/fix-agents-view-return-focus.md b/.changes/fix-agents-view-return-focus.md
deleted file mode 100644
index 3703ad396f..0000000000
--- a/.changes/fix-agents-view-return-focus.md
+++ /dev/null
@@ -1 +0,0 @@
-- Agents view: leaving a session now returns the focus to that session's row, even after `/switch`, `/new`, `/fork`, or `/resume` in the chat, and when the row loads late.
diff --git a/.changes/fix-open-family-registration-race.md b/.changes/fix-open-family-registration-race.md
deleted file mode 100644
index 009c4e5cef..0000000000
--- a/.changes/fix-open-family-registration-race.md
+++ /dev/null
@@ -1 +0,0 @@
-- Reopening a saved session now publishes its passive subagents before the first Agents view snapshot, even when worker registration is delayed or the opening client disconnects.
diff --git a/.changes/idle-spawn-prompt.md b/.changes/idle-spawn-prompt.md
deleted file mode 100644
index 95c635d5d0..0000000000
--- a/.changes/idle-spawn-prompt.md
+++ /dev/null
@@ -1 +0,0 @@
-- A subagent spawned while its parent has no turn running (for example a factory run resumed between turns) now gets its task prompt at once instead of up to 60 s later. A parent turn aborted by a manual compaction or tree move no longer leaves later spawns waiting for that bound either.
diff --git a/.changes/late-grandchild-attribution.md b/.changes/late-grandchild-attribution.md
deleted file mode 100644
index 37725c58b3..0000000000
--- a/.changes/late-grandchild-attribution.md
+++ /dev/null
@@ -1 +0,0 @@
-- A parent session's usage totals now include a grandchild's spend that lands after the parent already read the child's spawning turn. Before, that spend was dropped from the parent's totals, so long-running grandchildren were undercounted.
diff --git a/.changes/prime-inference-null-specs-warning.md b/.changes/prime-inference-null-specs-warning.md
deleted file mode 100644
index 354659532b..0000000000
--- a/.changes/prime-inference-null-specs-warning.md
+++ /dev/null
@@ -1 +0,0 @@
-- Prime Inference models dropped from the model catalog because the server sent no specs for them are now named in a log warning.
diff --git a/.changes/refine-headers.md b/.changes/refine-headers.md
deleted file mode 100644
index 90db5d1216..0000000000
--- a/.changes/refine-headers.md
+++ /dev/null
@@ -1 +0,0 @@
-- `/refine` and automatic refinement work again with Prime Inference team billing: they sent no team header and failed with "Refiner did not return a JSON object".
diff --git a/.changes/rlm-import-surface.md b/.changes/rlm-import-surface.md
deleted file mode 100644
index c516048162..0000000000
--- a/.changes/rlm-import-surface.md
+++ /dev/null
@@ -1 +0,0 @@
-- Kernel: `rlm.factory.run(...)`, `rlm.factory.help()`, and every other `rlm` namespace call keep working after a cell runs `import rlm`, and `dir(rlm.harness)` now lists the harness methods.
diff --git a/.changes/tool-execution-end-iserror.md b/.changes/tool-execution-end-iserror.md
deleted file mode 100644
index d389a817a7..0000000000
--- a/.changes/tool-execution-end-iserror.md
+++ /dev/null
@@ -1 +0,0 @@
-- `tool_execution_end` events on the RPC stream and in `--mode json` print output carry `isError` again, as the TypeScript version emitted them.
diff --git a/.changes/user-bash-signal-exit-code.md b/.changes/user-bash-signal-exit-code.md
deleted file mode 100644
index de2d142b89..0000000000
--- a/.changes/user-bash-signal-exit-code.md
+++ /dev/null
@@ -1 +0,0 @@
-- A `!` command or a model `bash` tool call whose shell dies from a signal (for example a crash, an OOM kill or `kill`) now reports 128 + the signal number as its exit code, so the model is told it failed, like the TypeScript version.
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0b75831865..e4c0fa8f83 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -3,6 +3,44 @@
Release notes for Prime Agent. Historical TypeScript release notes are
recorded in the [TypeScript changelog](https://github.com/PrimeIntellect-ai/prime-agent/blob/v0.9.8/packages/coding-agent/CHANGELOG.md).
+## [0.10.2] - 2026-10-28
+
+- The static usage layer no longer tells agents to await commands before ending the turn: since #3580 daemonless sessions also wake on a queued `[bash-done]` turn, every session kind has completion notifications, and the no-polling rule (end the turn and collect the result on the done notification) is the one turn-end contract.
+- Agents view: leaving a session now returns the focus to that session's row, even after `/switch`, `/new`, `/fork`, or `/resume` in the chat, and when the row loads late.
+- `tool_execution_end` events on the RPC stream and in `--mode json` print output carry `isError` again, as the TypeScript version emitted them.
+- Azure OpenAI Responses requests no longer fail with 404 on the v1 API: they now post to `<resource>/openai/v1/responses?api-version=v1` with the deployment as the request's `model`, instead of a `/deployments/<name>/responses` route Azure does not serve. A different `AZURE_OPENAI_API_VERSION` (for example `preview`, or a dated version against a custom `/openai` base URL) is sent as the `api-version` query instead, and an empty one falls back to `v1`.
+- A `!` command or a model `bash` tool call whose shell dies from a signal (for example a crash, an OOM kill or `kill`) now reports 128 + the signal number as its exit code, so the model is told it failed, like the TypeScript version.
+- ACP: a failed `session/prompt` now carries `kind`, `category` and `retryable` in its error data next to `details`, so a host can tell an exhausted model connection (`kind: "model_transport"`, the only retryable kind) from a context overflow (`context_limit`), an authentication or rate-limit error, or a harness failure.
+- The bundled websearch skill is now offered only when a Serper API key is set up (saved through `/mcp`, or `SERPER_API_KEY`), instead of being advertised on installs where every call fails; `bundledSkills.websearch` in settings still turns it on or off explicitly. A key saved in `/mcp` activates the skill in new sessions (`/new`).
+- A `python_repl` cell that runs longer than the new `pythonCellTimeoutMs` setting (default 60 seconds, `0` disables) now gets a note on stderr that points long shell work to `bash()` and long Python work to smaller cells. The cell is not stopped and keeps its result. Sessions without the daemon add no note.
+- When a running subagent is cancelled, deleted with `rlm.delete_subagent`, or stopped because its parent session is killed, the parent's semantic-edge ledger now records a `child_cancelled` line with the reason (`user_cancel`, `parent_delete` or `close:killed`) and no `child_returned` line beside it. A daemon restart or update records no such line.
+- A PageUp or wheel scroll that reaches the top of the loaded transcript while older messages are still loading is no longer lost: the view scrolls once the older page arrives.
+- `@` file completion is now instant: the project's files are indexed in the background when the TUI starts, suggestions come straight from that index, and the menu opens in the same frame as the keystroke. Large file changes (such as moving a big directory) no longer block typing. Which files match, and their order, is unchanged.
+- Requests to Gemini models through the Google and Vertex AI providers no longer fail with `400 INVALID_ARGUMENT`: the system prompt is now sent in the content format Google's API requires instead of as a plain string.
+- MCP OAuth login no longer fails when a server's advertised authorization server and its metadata `issuer` differ only by a trailing slash (for example TickTick).
+- Rows written around a compaction (a queued prompt, a rename or refinement notice) no longer show twice in the transcript. A steer or follow-up queued while you navigate the session tree or fork now waits until that finishes and runs on the branch you end up on, as the TypeScript version did.
+- Restoring a kernel snapshot no longer reopens saved file handles: a leftover handle opened for writing (even a closed one) used to empty its file on every restart. Restored file objects now come back closed, keeping their name and mode; open the file again to use it.
+- On openai-completions and Mistral providers, a turn whose tool calls were streamed without a function name no longer ends silently: the call gets a `Tool unknown_tool not found` error so the model can retry, and a nameless call next to named calls is dropped. Sessions that already stored such a call no longer fail every later request with `400 ... missing a function name`.
+- A subagent whose final model call fails with a provider error (after retries) now settles as `error` with the provider's message in `rlm.collect`, `rlm.list_subagents` and the factory, instead of `done` with no answer. If it had not replied yet, the parent gets a failure notice.
+- Requests through openai-completions, openai-responses and Mistral no longer send two `Authorization` headers when the model or the caller sets one (for example a custom provider with `authHeader: true`), so strict servers stop rejecting them.
+- A cost a provider reports in `usage.cost` is now used for any OpenAI-compatible provider, not only OpenRouter, so gateways in front of OpenRouter such as LiteLLM no longer record $0.
+- `prime-agent mcp add --bearer-token-env-var` now writes the `bearerTokenEnvVar` settings key the MCP engine reads, so HTTP MCP servers added this way send their bearer token instead of connecting without auth. Settings already written with `bearer_token_env_var` keep working.
+- Claude Haiku 5.x models on Anthropic and Bedrock now use adaptive thinking with an effort level instead of budget-based thinking, which the API rejects with a 400; on Bedrock, turning thinking off now disables it explicitly. Haiku 4.5 keeps budget-based thinking.
+- Prompt caching on openai-codex works again: every request in a session carries the same session id, which now stays stable across resumes and daemon restarts, so Codex WebSocket reuse and cached input work like the TypeScript version. Other providers also send the session id where the TypeScript version did (for example `prompt_cache_key` on OpenAI and Azure, and Mistral's affinity header), and side questions reuse the conversation's cache.
+- Bedrock works again: profiles that use `credential_process` now get credentials, requests are signed the way AWS expects instead of being rejected, and streamed replies are no longer empty. `AWS_CONFIG_FILE` and `AWS_SHARED_CREDENTIALS_FILE` are honored for credentials and region, and Claude 5 and newer models on Bedrock get prompt caching.
+- Fresh installs now ship the live model and MCP catalog, instead of a test snapshot that stayed listed until the first catalog refresh.
+- Gemini 3 and newer models on the Google and Vertex AI providers (for example `gemini-4-argon` and `gemini-flash-latest`) now send `thinkingLevel` instead of `thinkingBudget`, so the thinking levels actually differ, and thinking off sends the model's lowest level instead of a zero budget some models reject. Gemini 2.x keeps `thinkingBudget`.
+- On Anthropic-compatible endpoints that return thinking without a signature (such as MiMo, Kimi and DeepSeek), earlier thinking is now passed back as thinking instead of as visible text, so the model no longer echoes its reasoning more each turn. Claude models drop unsigned thinking, which they reject; the new `requiresThinkingSignature` compat flag overrides the per-model default.
+- A session with a large Python namespace no longer overflows its context after repeated compactions: each `[python-state]` notice lists at most 200 names or 4 KB (ending with `+N more`), and only the newest notice is kept in the model's context.
+- OpenCode and OpenCode Go requests carry the `x-opencode-session` header with the conversation id again, including compaction, refinement and branch-summary calls. Those three calls now send the session id to every provider for prompt caching, as the TypeScript version did, but always over SSE instead of the Codex WebSocket.
+- Prime Inference models dropped from the model catalog because the server sent no specs for them are now named in a log warning.
+- Kernel: `rlm.factory.run(...)`, `rlm.factory.help()`, and every other `rlm` namespace call keep working after a cell runs `import rlm`, and `dir(rlm.harness)` now lists the harness methods.
+- Factory machines can loop back into an entry state: an entry state may now declare inputs as long as all of them are optional. Its first entry binds the null sentinel, and a later entry through a transition binds the real output, so an investigate/verify loop no longer needs a duplicate re-investigate state. A required input on an entry state is still rejected, with a message saying it must be optional.
+- A parent session's usage totals now include a grandchild's spend that lands after the parent already read the child's spawning turn. Before, that spend was dropped from the parent's totals, so long-running grandchildren were undercounted.
+- Reopening a saved session now publishes its passive subagents before the first Agents view snapshot, even when worker registration is delayed or the opening client disconnects.
+- A subagent spawned while its parent has no turn running (for example a factory run resumed between turns) now gets its task prompt at once instead of up to 60 s later. A parent turn aborted by a manual compaction or tree move no longer leaves later spawns waiting for that bound either.
+- `/refine` and automatic refinement work again with Prime Inference team billing: they sent no team header and failed with "Refiner did not return a JSON object".
+
## [0.10.1] - 2026-10-21
- A corrupted line in the RLM spawn ledger is now skipped and logged instead of failing every ledger read, so session listings and new subagent spawns keep working.
diff --git a/Cargo.lock b/Cargo.lock
index c43032a564..5e7bf30988 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1441,7 +1441,7 @@ checksum = "7c87def4c32ab89d880effc9e097653c8da5d6ef28e6b539d313baaacfbafcbe"
[[package]]
name = "pa-agent"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"serde",
@@ -1452,7 +1452,7 @@ dependencies = [
[[package]]
name = "pa-ai"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"base64 0.22.1",
@@ -1487,7 +1487,7 @@ dependencies = [
[[package]]
name = "pa-cli"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"base64 0.22.1",
@@ -1517,7 +1517,7 @@ dependencies = [
[[package]]
name = "pa-core"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"base64 0.22.1",
@@ -1557,7 +1557,7 @@ dependencies = [
[[package]]
name = "pa-daemon"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"filetime",
@@ -1584,7 +1584,7 @@ dependencies = [
[[package]]
name = "pa-models"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"hex",
"hmac",
@@ -1604,7 +1604,7 @@ dependencies = [
[[package]]
name = "pa-perf"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"clap",
@@ -1620,7 +1620,7 @@ dependencies = [
[[package]]
name = "pa-telemetry"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"reqwest",
@@ -1635,7 +1635,7 @@ dependencies = [
[[package]]
name = "pa-tui"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"base64 0.22.1",
@@ -1663,7 +1663,7 @@ dependencies = [
[[package]]
name = "pa-types"
-version = "0.10.1"
+version = "0.10.2"
dependencies = [
"anyhow",
"libc",
diff --git a/Cargo.toml b/Cargo.toml
index c8851277e7..bbbf08ff95 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -20,7 +20,7 @@ members = [
crossterm = { path = "vendor/crossterm" }
[workspace.package]
-version = "0.10.1"
+version = "0.10.2"
edition = "2021"
license = "MIT"