* fix(llm-core): settle result() when a stream ends without a terminal event
EventStream.end() with no argument and no prior terminal event left the final
result promise pending forever; any consumer awaiting result() dead-ended
silently for the whole run budget (the agent loop's fall-through in
streamAssistantResponse awaits exactly that promise). All in-repo pumps push
terminals or pass end(result), so this only fires on producer contract
violations — which previously hung and now reject loudly with a named error.
A pre-attached catch keeps iterate-only consumers free of unhandled
rejections. Also dedupe the agent loop's two identical final-message blocks
into one finalizeAssistantMessage helper (net-negative production delta).
Follow-up from #120426's turn-liveness investigation.
* fix(llm-core): void the intentional rejection pre-catch
ClawSweeper P1: the pre-attached no-op catch returns a Promise that would
float under typescript/no-floating-promises.
* fix(agent): apply steering before unstarted tools
Restore steering checkpoints before sequential tool launches and before parallel batch launch. Preserve paired synthetic tool results, async callback compatibility, and Code Mode outcome handling.
* fix(agent): delay tool loop admission commits
Commit loop-detection history only for calls crossing the final launch checkpoint. Release steering-skipped markers, add repeated-steer coverage, and align remaining steering contract text.
* fix(agent): keep tool admission lifecycle internal
Attach delayed admission callbacks through the private internal-hooks seam so steering history remains correct without widening the public Agent Core or Plugin SDK contract.
* fix(agent): preserve steering API contracts
Keep public steering callbacks Promise-based and protocol error kinds unchanged. Use private synchronous draining and structured skip details to retain launch-boundary behavior without API or generated protocol drift.
* test(gateway): use canonical steering fixture config
Use keyed agent entries in the real gateway steering harness so current main does not migrate the fixture during startup.
* fix(agent): remove unused lifecycle re-export
* fix(agent): gate tool launch after wrapper preflight
Split OpenClaw tool execution into private prepare and launch phases so steering is checked after policy, approval, validation, and reconciliation but before the original side effect. Preserve final arguments, voice grants, loop admission, context wrappers, and direct tool execution.
* fix(agent): preserve steering callback receiver
Invoke public steering callbacks with their AgentLoopConfig receiver and cover method-style implementations that read config-owned queue state.
* [AI] fix(agents): emit aborted tool results for skipped tool calls on mid-turn abort
When an abort fires mid-batch in executeToolCalls (after the assistant
message with tool_use is committed but before all tool_results are
written), the sequential and parallel dispatch loops break out and skip
the remaining tool calls. The committed assistant turn retains N tool_use
blocks but only M < N tool_results land in context.messages, leaving
orphaned tool_use that corrupts retries/continuation and triggers provider
400 errors on providers that do not synthesize missing results
(allowSyntheticToolResults=false, e.g. openai-completions/DeepSeek).
Emit aborted tool results (createErrorToolResult("Operation aborted")) for
the skipped tail in both executeToolCallsSequential and
executeToolCallsParallel so every tool_use keeps a paired tool_result.
This complements the existing write-side guard (which only covers
synthetic-enabled providers) and the persisted replay repair.
The aborted tail outcomes are routed through finalizeToolCallOutcome (via
a shared finalizeAbortedToolCall helper) so config.afterToolOutcome hooks
(audit, redaction, metadata, error-normalization) observe these skipped
calls just like every immediate or executed outcome, instead of bypassing
the outcome contract. Regression tests assert afterToolOutcome fires for
every skipped call in both dispatch modes.
Fixes#116379
Co-Authored-By: Maas <noreply@anthropic.com>
* [AI] fix(agents): emit tool_execution_start before aborted end for skipped calls
The abort-tail backfill added in #116379 emits tool_execution_end (and a
paired tool_result) for tool calls the dispatch loop never reached, but it
skipped the matching tool_execution_start. Channel/client subscribers that
pair start→end events received an end for an unknown tool-call id during
abort recovery.
Emit tool_execution_start for each skipped call before its aborted end/result,
mirroring the start event every dispatched (including immediate non-executed)
call already emits. Covers both sequential and parallel dispatch, with
regression assertions that every skipped call has a start before its end and
that start/end counts stay paired.
Co-Authored-By: Maas <noreply@anthropic.com>
* fix(agents): complete aborted tool tails safely
Fixes#116379
---------
Co-authored-by: Maas <noreply@anthropic.com>
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
* test(process): reproduce Windows taskkill process-tree leak
* fix(process): stop leaked Windows child process trees
Escalate only when Windows taskkill reports that graceful process-tree termination failed. Preserve awaited taskkill completion, grace-period fallback, one-shot signaling, and PID-reuse protection.
Closes#110789
Supersedes #112202
Co-authored-by: Mohammed Alkindi <alkndymhmd692@gmail.com>
---------
Co-authored-by: Mohammed Alkindi <alkndymhmd692@gmail.com>
* feat(anthropic): add Claude Opus 5 model support
Wire claude-opus-5 through the Claude 5 contract seams: adaptive-by-default
thinking with the full low..max effort range, default-sampling and prefill
stripping, streaming refusal contract, model-bound thinking replay, and 1M/128k
catalog metadata across anthropic, claude-cli, Vertex, Bedrock, and Mantle.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(anthropic-vertex): normalize explicit Vertex Opus 5 model rows
Extend normalizeAnthropicVertexResolvedModel so user-configured Vertex Opus 5
rows regain reasoning, image input, 1M/128k limits, and the native thinking
map; update the live-model priority expectation and regenerate docs_map.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
* fix(infra): harden network client errors, session short ids, and extension load caching
* fix(infra): preserve injected dispatcher seams and drop the uncached extension loader
* fix(kill-tree): verify process group leader before group kill to prevent gateway SIGTERM (#76259)
- Add isProcessGroupLeader() to killProcessTree/signalProcessTree: ps -p <pid> -o pgid= primary check with /proc/<pid>/stat fallback on Linux. Group kill only when the PID is its own process group leader; non-leaders fall back to single-pid kill, preventing accidental gateway SIGTERM when a non-detached child shares the gateway's process group.
- Propagate detached: true to all detached-spawn cleanup callers (exec-termination, agent-bundle LSP, mcp-stdio, bash, supervisor pty, agent-core nodejs) so detached group cleanup survives leader exit.
- Gateway/daemon cleanup paths (schtasks, restart-health) keep the leader-checked default (detached omitted).
Closes#76259
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor(process): tighten process-group ownership checks
* refactor(daemon): split restart diagnostics
* refactor(daemon): isolate restart health types
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Peter Steinberger <steipete@gmail.com>
* feat(tooling): extend strict-ratchet lane to all remaining leaf packages
* fix(packages): burn down indexed-access debt in ai and agent-core
All 196 noUncheckedIndexedAccess errors fixed behavior-identically
(iteration, .at()/slices, DataView byte math, guarded queue peeks) plus
cleanup finds: a sparse-tool diagnostic bug, deduplicated Responses
call/item ID parsing with an honest string|undefined splitter, and a
clearer local failure for redacted-thinking blocks missing signatures.
Removes all 17 remaining non-null assertions across ai, gateway-client,
and llm-core so the assertion ban aligns 1:1 with the ratchet lane.
speech-core joins memory-host-sdk as structurally deferred (imports
src/** via plugin-sdk path mappings).
* fix(agent-core): keep file info basename on Windows paths
* fix(agent-core): use platform path basenames
* chore: sync PR branch before changelog
* docs(changelog): credit Windows file metadata fix
* fix(agent-core): use platform path basenames
* test(agent-core): isolate Windows shell discovery
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com>
Co-authored-by: Peter Steinberger <peter@steipete.me>
* fix(agent-core): handle stdout/stderr stream errors in harness exec
Register error handlers on stdout and stderr streams in
NodeExecutionEnv.exec() to prevent uncaught exceptions when
a pipe breaks (e.g. EPIPE after child process exit).
Without these listeners, Node.js throws an uncaught exception
that crashes the entire agent-core process.
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(agent-core): address review — hoist-safe mock and curly lint
- Use vi.hoisted pattern for spawnMock to fix ReferenceError in CI
- Wrap one-line if guard in braces per repo curly rule
Co-Authored-By: Claude <noreply@anthropic.com>
* chore: keep release notes in PR context
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Peter Steinberger <steipete@gmail.com>
The active user turn was decorated at request-build time with an inbound
metadata block (Conversation info / reply-target / sender / forwarded /
chat-history), but the session store persists the bare text and the LLM boundary
strips that block from historical replay. So a user message's serialized bytes
changed retroactively — decorated as the active turn, bare as history — on every
turn, invalidating any prefix/exact-match provider prompt cache from that point
onward, every turn. (#90811 fixed the timestamp half of this same asymmetry.)
Rework of the earlier "strip the active turn" approach (which lost model-visible
context — ClawSweeper's P1) into one that keeps the context AND recovers the
cache cost:
- Route current-turn inbound metadata out of the user prompt and into the hidden
runtime-context custom message (the mechanism from openclaw#89428), and
relocate that carrier to the ABSOLUTE TAIL of the wire request (after the
active user turn and any tool-call scaffolding). The user turn stays bare and
byte-identical in both the active and historical positions; the volatile
carrier vanishes next turn exactly where the assistant reply begins anyway, so
the request is an append-only prefix-extension through the active user turn.
- A durable marker (UserMessage.runtimeContextCarrier, set by convertToLlm from
the carrier's details) lets the Anthropic SDK transport, the managed Anthropic
transport (anthropic-transport-stream / anthropic-payload-policy), and
OpenAI-completions skip the carrier when selecting the deepest cache_control
breakpoint, keeping the anchor on the last stable user turn.
Storage stays BARE (invariant preserved); runtime-only turns (room events) keep
their existing inline behavior, which is byte-stable because room-event/system
context is not strip-eligible; legacy bare-stored sessions are unchanged.
Reviewed with adversarial gpt-5.5/codex sweeps (correctness, byte-identity,
compatibility, test adequacy): found + fixed the managed Anthropic transport
cache-anchor path and added the missing OpenAI-completions/managed-transport
breakpoint tests; a theoretical runtime-only mutation was verified unreachable
and pinned with a test. Final sweep clean.
Co-authored-by: openclaw#89428 (runtime-context handoff approach)
Compaction failure and timeout recovery can no longer leave a session transcript assistant-last and wedged after restoring pre-compaction state.
Transcript-continuation failures now carry a typed agent-core error code and no longer demote the active model through model fallback scoring.
Surface: embedded agent runner compaction recovery, model fallback classification, packages/agent-core.
Refs #100312. Includes regression coverage for #99943.
Centralized managed worktrees under <state-dir>/worktrees/<repo-fingerprint>/<name>
with branch-per-task (openclaw/<name>), .worktreeinclude provisioning, an optional
.openclaw/worktree-setup.sh repo hook, and a SQLite registry in the shared state DB.
Removal always snapshots the tree (untracked included, gitignored excluded) to
refs/openclaw/snapshots/<id>; restore rebuilds the branch at the original commit with
the snapshot content as uncommitted state. Lossless run-end cleanup, 7-day idle GC for
run-owned worktrees (manual exempt), orphan reconciliation, 30-day snapshot retention.
Surfaces: worktrees.* gateway RPC (operator.admin mutations), openclaw worktrees CLI,
Control UI page, plugin-SDK facade + Workboard kind:"worktree" materialization.
E2E-verified on Testbox: full create->work->remove->restore->gc lifecycle.
Peter Steinberger