Require routable ClawSweeper commands before fast ack

[brokemac79]

Summary

Fixes #232.

This makes the fast-ack paths reject comments that only mention @clawsweeper inline inside prose, unless the same comment is actually routable by the command parser.

Changes:

• gates the deterministic comment webhook on parseCommand(...) before posting the visible fast ack
• extracts the shared ClawSweeper command-line detector used by both the router parser and hosted dashboard worker
• replaces the hosted dashboard worker's loose mention regex with that shared parser-compatible detector
• keeps valid command forms working, including an own-line mention and the @clawsweeper then next-line re-review pattern used by issue authors
• adds regression coverage for inline mentions versus real command-line mentions
• normalizes a few workflow test reads so CRLF workflow files do not break unrelated assertions

Testing

• pnpm run build:all
• pnpm run lint
• node --test test\\repair\\comment-router-core.test.ts test\\repair\\comment-webhook.test.ts test\\dashboard-worker.test.ts
• node --test --test-name-pattern "sweep workflow_dispatch input count|sweep review continuations|repair workflows preserve" test\\clawsweeper.test.ts
• pnpm exec oxfmt --check src/repair/comment-command-text.ts src/repair/comment-router-core.ts dashboard/worker.ts

Note: I also tried pnpm run check locally on Windows earlier. The relevant build/lint/regression parts pass, but the full suite still hits unrelated Windows/local-environment failures around spawning codex and POSIX file-mode assertions. A repository-wide format:check also reports existing unrelated formatting noise on this Windows checkout, while the touched files pass targeted format check.

[clawsweeper]

Codex review: needs real behavior proof before merge. Reviewed May 30, 2026, 3:33 PM ET / 19:33 UTC.

Summary
The branch adds a shared command-line detector, uses it to gate standalone and hosted fast-ack classification, and adds regression coverage plus CRLF-tolerant workflow test reads.

Reproducibility: yes. source-reproducible: current main's loose fast-ack regex matches the reported inline mention while parseCommand only accepts command-shaped lines. I did not run live webhook delivery.

Review metrics: 2 noteworthy metrics.

• Fast-ack classifiers: 2 paths changed. Both the standalone webhook and hosted dashboard worker can affect live ClawSweeper command intake.
• Changed surface: 8 files, +217/-97. The patch is bounded but touches source, hosted worker code, and regression tests.

Merge readiness
Overall: 🧂 unranked krab
Proof: 🧂 unranked krab
Patch quality: 🐚 platinum hermit
Result: blocked until real behavior proof is added.

Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch.

Rank-up moves:

• [P1] Add redacted live output from a real signed issue_comment webhook or hosted-worker run showing inline prose rejected and own-line commands accepted.
• After adding proof, update the PR body so ClawSweeper re-reviews automatically, or ask a maintainer to comment @clawsweeper re-review.

Proof guidance:

• [P1] Needs real behavior proof before merge: No redacted live signed-webhook or hosted-worker output is present; the contributor should add terminal/log/live output with private details redacted and update the PR body to trigger re-review.

Risk before merge

• [P1] No redacted live webhook or hosted-worker output is supplied yet, so signed issue_comment delivery and visible fast-ack behavior are only covered by tests and source inspection.
• [P1] The diff changes both deterministic and hosted fast-ack command classifiers; a production regression could suppress valid commands or acknowledge non-command prose.

Maintainer options:

1. Require live command-routing proof (recommended)
   Ask for redacted output from a real signed issue_comment webhook or hosted-worker run showing inline prose rejected and own-line commands accepted before merge.
2. Accept source-level coverage risk
   Maintainers can merge based on source inspection and regression tests if they are comfortable owning the remaining live-runtime uncertainty.

Next step before merge

• [P1] Needs contributor-supplied real behavior proof and normal maintainer review; no narrow code repair is currently identified.

Security
Cleared: The diff changes command parsing and tests without adding dependencies, secrets handling, permissions, downloaded code, or package-resolution changes.

Review details

Best possible solution:

Merge a narrow parser-compatible fast-ack gate after redacted real webhook or hosted-worker proof shows inline prose is ignored and own-line re-review commands still dispatch.

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

Yes, source-reproducible: current main's loose fast-ack regex matches the reported inline mention while parseCommand only accepts command-shaped lines. I did not run live webhook delivery.

Is this the best way to solve the issue?

Yes, the shared parser-compatible command-line detector is the narrow maintainable fix. The remaining blocker is proof that the same behavior works in a real webhook or hosted-worker run.

AGENTS.md: found and applied where relevant.

Codex review notes: model gpt-5.5, reasoning high; reviewed against 83cb99d7a283.

Label changes

Label justifications:

• P2: This is a bounded automation bug fix with limited blast radius, but it affects visible command acknowledgement and routing.
• merge-risk: 🚨 automation: The PR changes live fast-ack command classification, where a regression could misroute, suppress, or falsely acknowledge ClawSweeper commands.
• rating: 🧂 unranked krab: Overall readiness is 🧂 unranked krab; proof is 🧂 unranked krab and patch quality is 🐚 platinum hermit.
• status: 📣 needs proof: The PR needs real behavior proof before ClawSweeper can clear the contributor ask. Needs real behavior proof before merge: No redacted live signed-webhook or hosted-worker output is present; the contributor should add terminal/log/live output with private details redacted and update the PR body to trigger re-review.

Evidence reviewed

What I checked:

• AGENTS.md policy read: Repository policy was read fully and applied: ClawSweeper reviews should be conservative, evidence-backed, automation-safe, and use pnpm run check before handoff for code changes. (AGENTS.md:1, 83cb99d7a283)
• Current main has the reported standalone fast-ack mismatch: On current main, COMMAND_PATTERN can match a ClawSweeper mention in prose, while parseCommand(...) is computed but not used to reject a null parse before accepted maintainer comments proceed to fast ack. (src/repair/comment-webhook.ts:174, 83cb99d7a283)
• Current main has the same hosted-worker loose gate: The hosted dashboard worker currently uses CLAWSWEEPER_COMMAND_PATTERN directly before fast ack, so it can accept mention-shaped prose without checking the router grammar. (dashboard/worker.ts:508, 83cb99d7a283)
• Router grammar accepts command-shaped lines: parseCommand scans lines and only accepts slash or mention commands when the line itself matches the command grammar, which is the behavior the PR is aligning fast ack with. (src/repair/comment-router-core.ts:1025, 83cb99d7a283)
• PR gates both fast-ack paths on parser-compatible command text: The PR adds extractClawSweeperCommandLine(...) and commandTextForClawSweeperFastAck(...), then uses the helper in dashboard/worker.ts and adds a null-parse rejection in comment-webhook.ts. (src/repair/comment-command-text.ts:17, c636d95b5da4)
• Regression tests cover the central command shapes: The PR adds tests rejecting inline prose mentions before visible ack, keeping own-line mention commands accepted, and checking the hosted worker false-positive case. (test/repair/comment-webhook.test.ts:39, c636d95b5da4)

Likely related people:

• Peter Steinberger: Blame shows 9efa568 introduced the standalone webhook regex, hosted-worker regex, and router parser surface that this PR changes. (role: introduced affected surface; confidence: high; commits: 9efa568da092; files: src/repair/comment-webhook.ts, dashboard/worker.ts, src/repair/comment-router-core.ts)
• Tak Hoffman: Recent commits changed the command webhook and router path, including the current parsedCommand use for stale closed PR command handling. (role: recent repair-command contributor; confidence: medium; commits: 9b0b0b391733, c01469876a85, 8db024ba0503; files: src/repair/comment-webhook.ts, src/repair/comment-router-core.ts, test/repair/comment-webhook.test.ts)
• brokemac79: Prior merged history on current main changed re-review command parsing and duplicate fast-ack handling in the same command-routing area, beyond authoring this PR. (role: recent command-routing contributor; confidence: high; commits: 485f71abc664, 556268dff3d9; files: src/repair/comment-router-core.ts, dashboard/worker.ts, test/repair/comment-router-core.test.ts)

What the crustacean ranks mean

• 🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
• 🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
• 🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
• 🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
• 🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
• 🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
• 🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

How this review workflow works

• ClawSweeper keeps one durable marker-backed review comment per issue or PR.
• Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
• A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
• PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
• Maintainers can also comment @clawsweeper review to request a fresh review only.
• Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
• Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
• Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

[brokemac79]

@clawsweeper re-review

Updated to extract the shared command-line detector so the router parser and hosted dashboard worker no longer maintain separate command-shape regexes.

[clawsweeper]

🦞🧹
ClawSweeper re-review requested.

I asked ClawSweeper to review this item again.
Action: item re-review queued (workflow sweep.yml, event repository_dispatch).
Result: the existing ClawSweeper review comment will be edited in place when the review finishes.

Re-review progress:

• State: Complete
• Detail: The targeted re-review finished, the durable review comment was updated, and the synced verdict was routed.
• Run: https://github.com/openclaw/clawsweeper/actions/runs/26692860171
• Updated: 2026-05-30T19:34:14.816Z

[Takhoffman]

@clawsweeper automerge

[clawsweeper]

🦞👀
ClawSweeper picked this up.

Command router queued. I will update this comment with the next step.