fix(active-memory): fast-fail stalled recall paths by clawsweeper[bot] · Pull Request #76183 · openclaw/openclaw

clawsweeper · 2026-05-02T17:38:28Z

Repair plan for #75761:

Rebase/refresh fix/active-memory-empty-search-fast-fail onto current main (9bedcff904dda8e153e1006b868b0f0532abecdf) and resolve the dirty merge state.
Keep the implementation limited to Active Memory runtime/tests and the existing changelog entry.
Address the current blockers: check-lint, check-additional-extension-bundled, check-additional, and the aggregate check failure.
Preserve the intended lifecycle: Active Memory still runs in before_prompt_build; useful recall still injects <active_memory_plugin>; clean zero-hit/unavailable Memory Search and hard timeout paths resolve as empty/timeout without injecting timeout boilerplate.
Keep QMD recall quality work in fix(memory): improve QMD recall for channel queries #75860 separate.

Validation before re-review:

pnpm exec oxfmt --check --threads=1 CHANGELOG.md extensions/active-memory/index.ts extensions/active-memory/index.test.ts
pnpm exec oxlint extensions/active-memory/index.ts extensions/active-memory/index.test.ts
pnpm test extensions/active-memory/index.test.ts
pnpm check:changed

After repair, run a fresh ClawSweeper/Codex review on the exact repaired head before any merge-capable applicator proceeds.

ClawSweeper 🐠 replacement reef notes:

Cluster: automerge-openclaw-openclaw-75761
Source PRs: fix(active-memory): fast-fail stalled recall paths #75761
Credit: Repair the existing PR branch for fix(active-memory): fast-fail stalled recall paths #75761 so source commit/PR credit for @codexGW is preserved in PR history.; If the changelog entry needs conflict repair, keep it as one single-line active-version user-facing fix entry; do not add forbidden Thanks attribution for @codex, @openclaw, @clawsweeper, or @steipete.; Keep fix(memory): improve QMD recall for channel queries #75860 credited and separate; do not fold QMD channel recall quality changes into this Active Memory timeout repair.
Validation: pnpm exec oxfmt --check --threads=1 CHANGELOG.md extensions/active-memory/index.ts extensions/active-memory/index.test.ts; pnpm exec oxlint extensions/active-memory/index.ts extensions/active-memory/index.test.ts; pnpm test extensions/active-memory/index.test.ts; pnpm check:changed
Repair fallback: To https://github.com/codexGW/openclaw.git
! [remote rejected] HEAD -> fix/active-memory-empty-search-fast-fail (refusing to allow a GitHub App to create or update workflow .github/workflows/ci.yml without workflows permission)
error: failed to push some refs to 'https://github.com/codexGW/openclaw.git'

fish notes: model gpt-5.5, reasoning high; reviewed against 3a55e61.

clawsweeper · 2026-05-02T17:39:44Z

Codex review: passed.

Summary
This PR adds Active Memory transcript polling to fast-fail terminal zero-hit or unavailable recall tool results, filters timeout boilerplate, extends focused regressions, and adds a changelog fix entry.

Reproducibility: yes. The PR includes focused regressions that reproduce terminal zero-hit search, unavailable search, non-empty details.results with debug.hits: 0, memory_get misses, and timeout boilerplate behavior.

Next step before merge
No repair lane is needed; this automerge-opted PR has no blocking review findings and should proceed through exact-head CI and mergeability gates.

Security
Cleared: The diff only changes Active Memory runtime logic, colocated tests, and a changelog entry; it does not alter dependencies, workflows, package scripts, permissions, or secret handling.

Review details

Best possible solution:

Merge this scoped Active Memory fix once exact-head CI and automerge gates are green, leaving the QMD recall-quality changes to the separate open follow-up.

Do we have a high-confidence way to reproduce the issue?

Yes. The PR includes focused regressions that reproduce terminal zero-hit search, unavailable search, non-empty details.results with debug.hits: 0, memory_get misses, and timeout boilerplate behavior.

Is this the best way to solve the issue?

Yes. The repaired approach is narrow and plugin-owned: it observes the existing transcript/tool-result contract, preserves useful non-empty recall results, and avoids pulling separate QMD quality work into this timeout fix.

What I checked:

Current main behavior: Current main races the embedded recall subagent only against the timeout promise, so it has no terminal memory-search watcher and still waits until timeout for clean empty/unavailable recall paths. (extensions/active-memory/index.ts:2315, 192e750035e7)
PR terminal result guard: The repaired head classifies terminal empty search only when the returned details.results array is empty when present, otherwise falling back to debug.hits === 0; this preserves non-empty returned results even when debug telemetry reports zero hits. (extensions/active-memory/index.ts:1600, e5ea3f1a7a4e)
PR race integration: The repaired head races terminalMemorySearchWatch.promise with the subagent and timeout, aborts the embedded run when terminal recall wins, persists an empty status, and resets the timeout circuit breaker. (extensions/active-memory/index.ts:2470, e5ea3f1a7a4e)
Memory Core result contract: Memory Core sets debug.hits from raw memory hits before merging supplement results into the final returned results array, which supports the repaired non-empty-results guard. (extensions/memory-core/src/tools.ts:310, 192e750035e7)
Regression coverage: The PR adds focused tests for zero-hit fast-fail, non-empty results with debug.hits: 0, unavailable search, memory_get miss non-terminal behavior, hard timeout, and timeout-boilerplate filtering. (extensions/active-memory/index.test.ts:2305, e5ea3f1a7a4e)
Validation state: The repair comment reports pnpm check:changed; the public check-runs API for the exact repaired head showed no failures at inspection time, with 68 success, 9 skipped, and 2 still in progress. (e5ea3f1a7a4e)

Likely related people:

steipete: Recent Active Memory timeout/setup-grace work, timeout-test stabilization, and adjacent Memory Core tool-config work touched the same recall/runtime surface. (role: recent maintainer; confidence: high; commits: 59449d7f19cc, d85d782a0a27, 8a8cc8dc9fca; files: extensions/active-memory/index.ts, extensions/active-memory/index.test.ts, extensions/memory-core/src/tools.ts)
Takhoffman: Prior merged Active Memory bundled recall/QMD work changed the same recall lifecycle, and the PR discussion shows Takhoffman re-enabled ClawSweeper automerge after an earlier pause. (role: adjacent owner; confidence: medium; commits: f256eeba431b, 885209ed0330, 4f00b769251d; files: extensions/active-memory/index.ts, extensions/memory-core/src/tools.ts)
vyctorbrzezowski: Merged work exposed memory tools to Active Memory recall runs, directly adjacent to this PR's memory_recall/memory_search/memory_get behavior. (role: adjacent owner; confidence: medium; commits: eabab1f64f9e; files: extensions/active-memory/index.ts, extensions/active-memory/index.test.ts)

Remaining risk / open question:

Exact-head CI still had two in-progress check runs at inspection time, so merge should remain gated on final CI and mergeability.

Codex review notes: model gpt-5.5, reasoning high; reviewed against 192e750035e7.

clawsweeper · 2026-05-02T17:41:47Z

🦞🦞
ClawSweeper merged this PR after the passing review.

Source: clawsweeper[bot]
Feedback: structured ClawSweeper verdict: pass (sha=e5ea3f1a7a4e84f55422c426b6471179d23b2c53)
Merge status: merged by ClawSweeper automerge
Merged at: 2026-05-02T22:23:03Z
Merge commit: ae82a39150f9

What merged:

This PR adds Active Memory transcript polling to fast-fail terminal zero-hit or unavailable recall tool results, filters timeout boilerplate, extends focused regressions, and adds a changelog fix entry.
Reproducibility: yes. The PR includes focused regressions that reproduce terminal zero-hit search, unavailab ... rch, non-empty details.results with debug.hits: 0, memory_get misses, and timeout boilerplate behavior.

Fixups included:

Included follow-up commit: fix(active-memory): fast-fail stalled recall paths
Included follow-up commit: fix(clawsweeper): address review for automerge-openclaw-openclaw-7576…
Included follow-up commit: fix(clawsweeper): reconcile automerge-openclaw-openclaw-75761 with ma…
Ran the ClawSweeper repair loop before final review.

The automerge loop is complete.

Automerge progress:

2026-05-02 22:11:35 UTC review requested repair [`3a55e615b3a1`](https://github.com/openclaw/openclaw/commit/3a55e615b3a14cc383c6c37dad690faad0bce3b2) (structured ClawSweeper marker: fix-required (finding=review-feedback sha=3a55e6...)
2026-05-02 17:41:46 UTC repair queued [`3a55e615b3a1`](https://github.com/openclaw/openclaw/commit/3a55e615b3a14cc383c6c37dad690faad0bce3b2) (autonomous) Run: https://github.com/openclaw/clawsweeper/actions/runs/25257889652
2026-05-02 18:11:42 UTC repair completed (no branch change) in 27m 40s Run: https://github.com/openclaw/clawsweeper/actions/runs/25257889652 source PR fix(active-memory): fast-fail stalled recall paths #76183 is paused by clawsweeper:human-review; refusing to mutate the PR branch
2026-05-02 22:07:46 UTC review queued [`3a55e615b3a1`](https://github.com/openclaw/openclaw/commit/3a55e615b3a14cc383c6c37dad690faad0bce3b2) (queued)
2026-05-02 21:57:25 UTC repair queued [`3a55e615b3a1`](https://github.com/openclaw/openclaw/commit/3a55e615b3a14cc383c6c37dad690faad0bce3b2) (autonomous) Run: https://github.com/openclaw/clawsweeper/actions/runs/25262875217
2026-05-02 22:19:07 UTC repair completed [`e5ea3f1a7a4e`](https://github.com/openclaw/openclaw/commit/e5ea3f1a7a4e84f55422c426b6471179d23b2c53) (branch updated) in 19m 15s Run: https://github.com/openclaw/clawsweeper/actions/runs/25262875217 initial automerge rebase is delegated to Codex repair
2026-05-02 22:19:07 UTC review queued [`e5ea3f1a7a4e`](https://github.com/openclaw/openclaw/commit/e5ea3f1a7a4e84f55422c426b6471179d23b2c53) (after repair)
2026-05-02 22:22:37 UTC review passed [`e5ea3f1a7a4e`](https://github.com/openclaw/openclaw/commit/e5ea3f1a7a4e84f55422c426b6471179d23b2c53) (structured ClawSweeper verdict: pass (sha=e5ea3f1a7a4e84f55422c426b6471179d23b2...)
2026-05-02 22:23:05 UTC merged [`e5ea3f1a7a4e`](https://github.com/openclaw/openclaw/commit/e5ea3f1a7a4e84f55422c426b6471179d23b2c53) (merged by ClawSweeper automerge)

Takhoffman · 2026-05-02T17:52:16Z

@clawsweeper stop

clawsweeper · 2026-05-02T17:53:10Z

🦞🦞
Got it. ClawSweeper will leave this item for human review.

I added clawsweeper:human-review and paused the automation trail until a maintainer asks again.

Takhoffman · 2026-05-02T21:53:30Z

@clawsweeper automerge

Co-authored-by: codexGW <9350182+codexGW@users.noreply.github.com>

…1 (1)

…in (1)

Summary: - This PR adds Active Memory transcript polling to fast-fail terminal zero-hit or unavailable recall tool results, filters timeout boilerplate, extends focused regressions, and adds a changelog fix entry. - Reproducibility: yes. The PR includes focused regressions that reproduce terminal zero-hit search, unavailab ... rch, non-empty `details.results` with `debug.hits: 0`, memory_get misses, and timeout boilerplate behavior. ClawSweeper fixups: - Included follow-up commit: fix(active-memory): fast-fail stalled recall paths - Included follow-up commit: fix(clawsweeper): address review for automerge-openclaw-openclaw-7576… - Included follow-up commit: fix(clawsweeper): reconcile automerge-openclaw-openclaw-75761 with ma… - Ran the ClawSweeper repair loop before final review. Validation: - ClawSweeper review passed for head e5ea3f1. - Required merge gates passed before the squash merge. Prepared head SHA: e5ea3f1 Review: openclaw#76183 (comment) Co-authored-by: clawsweeper <274271284+clawsweeper[bot]@users.noreply.github.com> Co-authored-by: codexGW <9350182+codexGW@users.noreply.github.com>

clawsweeper Bot added size: L clawsweeper:automerge Maintainer opted this PR into bounded ClawSweeper-reviewed automerge clawsweeper Tracked by ClawSweeper automation labels May 2, 2026

clawsweeper Bot mentioned this pull request May 2, 2026

fix(active-memory): fast-fail stalled recall paths #75761

Closed

openclaw-barnacle Bot added size: M and removed size: L labels May 2, 2026

clawsweeper Bot added the clawsweeper:human-review Needs maintainer review before ClawSweeper can continue label May 2, 2026

clawsweeper Bot removed the clawsweeper:human-review Needs maintainer review before ClawSweeper can continue label May 2, 2026

clawsweeper Bot and others added 4 commits May 2, 2026 22:13

fix(active-memory): fast-fail stalled recall paths

bcec726

Co-authored-by: codexGW <9350182+codexGW@users.noreply.github.com>

fix(active-memory): fast-fail stalled recall paths

714459a

fix(clawsweeper): address review for automerge-openclaw-openclaw-7576…

774ce25

…1 (1)

fix(clawsweeper): reconcile automerge-openclaw-openclaw-75761 with ma…

e5ea3f1

…in (1)

clawsweeper Bot force-pushed the clawsweeper/automerge-openclaw-openclaw-75761 branch from 3a55e61 to e5ea3f1 Compare May 2, 2026 22:19

clawsweeper Bot merged commit ae82a39 into main May 2, 2026
81 checks passed

clawsweeper Bot deleted the clawsweeper/automerge-openclaw-openclaw-75761 branch May 2, 2026 22:23

github-actions Bot mentioned this pull request May 2, 2026

📡 Upstream Digest — 2026-05-02 22:32 UTC curtismercier/openclaw-mods#750

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(active-memory): fast-fail stalled recall paths#76183

fix(active-memory): fast-fail stalled recall paths#76183
clawsweeper[bot] merged 4 commits intomainfrom
clawsweeper/automerge-openclaw-openclaw-75761

clawsweeper Bot commented May 2, 2026

Uh oh!

clawsweeper Bot commented May 2, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented May 2, 2026 •

edited

Loading

Uh oh!

Takhoffman commented May 2, 2026

Uh oh!

clawsweeper Bot commented May 2, 2026

Uh oh!

Takhoffman commented May 2, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Conversation

clawsweeper Bot commented May 2, 2026

Uh oh!

clawsweeper Bot commented May 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clawsweeper Bot commented May 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Takhoffman commented May 2, 2026

Uh oh!

clawsweeper Bot commented May 2, 2026

Uh oh!

Takhoffman commented May 2, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

clawsweeper Bot commented May 2, 2026 •

edited

Loading

clawsweeper Bot commented May 2, 2026 •

edited

Loading