Skip to content

build(deps): refresh compatible fixtures and repair telemetry audit - #295

Merged
steipete merged 9 commits into
mainfrom
chore/deps-refresh-20260830
Sep 22, 2026
Merged

steipete merged 9 commits into
mainfrom
chore/deps-refresh-20260830

Conversation

@steipete

@steipete steipete commented Aug 31, 2026 •

Copy link
Copy Markdown
Contributor

What Problem This Solves

Fixture fallback pins and committed dependency locks lag behind compatible stable packages. TencentDB's existing lock also retains a vulnerable Jaeger propagator that a normal compatible audit repair cannot update.

User Impact

Refresh 21 OpenClaw fixture pins to 2026.9.5, LightClawBot to 1.3.0, Weixin to 2.4.9, and Yuanbao to 2.18.3. Update compatible overrides, npm-generated locks, pnpm to 11.27.1, and pnpm/action-setup to 6.1.0. Preserve the promoted OpenClaw host, historical behavior targets, and external Git fixture pins. Builds on the exact reviewed runtime capture repairs in #350.

Discord's existing 2026.7.1 fallback manifest and lock remain byte-for-byte unchanged from main. Actual Discord 2026.9.5 needs atomic state-store observation/comparison methods absent from the pinned host and Inspector runtime. Its latest-track failure remains an explicit compatibility hold, and the incompatible standalone bump #347 is closed with evidence. A host promotion and matching Inspector runtime support should precede that update.

Dependency Boundaries

All changed locks use the repository's npm legacy-peer-deps procedure. Updated Weixin and LightClawBot locks exclude automatic OpenClaw peers. HAPI's explicit host dependency retains its published shrinkwrap: the full 309-entry graph is unchanged after removing six ineffective overrides. Its mutable Hono dependency advances to 4.13.8; a real npm ci passed.

TencentDB keeps the exact 1.0.1 artifact, registry URL, integrity, and SDK 0.218.0. A narrow stable-major Jaeger override selects 2.11.0. Normal audit repair retained Jaeger 2.7.1 and two high findings. The publisher-transition decision in #307 remains separate.

Observable Validation

The real SDK-configured propagator round trip after a clean install, with network exporters disabled, reports:

{"fixture":"1.0.1","sdkNode":"0.218.0","jaeger":"2.11.0","sdkConfiguredPropagator":"jaeger","traceRoundTrip":"pass","networkExporters":"none"}

The unchanged security checker ran against the exact tracked npm files, matching CI's non-recursive checkout. External submodule lockfiles remain separately owned upstream inputs, not modified or exempted by this PR.

$ npm run check:fixture-security
crabpot: no lock-refreshable critical/high fixture vulnerabilities

$ node scripts/run-static-suite.mjs --openclaw ./openclaw --policy dashboard --profile-runs 3 --plugin-inspector-smoke --openclaw-track latest --plugin-track latest
ℹ tests 435
ℹ pass 423
ℹ fail 0
ℹ skipped 12
contract coverage ok: 269 issues, 237 probes, 0 breakages
ci policy: pass; 0 fail, 1 warn, 7 pass
crabpot: static step 19/19 complete

All eleven repaired isolated lanes were then exercised together: HAPI, TencentDB, Matrix, Voice Call, Brave, Codex, Google Meet, diagnostics-otel, Nostr, WhatsApp, and Feishu.

$ CRABPOT_EXECUTE_ISOLATED=1 node scripts/execute-workspace-plan.mjs --fixture-set hapi-openclaw,memory-tencentdb,matrix,voice-call,brave-plugin,codex,google-meet,diagnostics-otel,nostr,whatsapp,feishu --allow-empty --continue-on-error --openclaw ./openclaw
$ node scripts/summarize-execution-results.mjs
execution results: 49 artifacts, 46 pass, 0 fail, 66 blocked
$ node scripts/check-ci-policy.mjs
ci policy: pass; 0 fail, 62 warn, 7 pass

The blockers remain explicit coverage gaps; HAPI/TencentDB preserve their raw known failures and Matrix/Voice Call require named runtime prerequisites. Unknown failures still fail policy. #350 records source and freshly packed Inspector smoke proof (59 fixtures, zero breakages, computed tool value, preserved Gateway rejection).

Windows CI exposed CRLF conversion of two captured stderr fixtures. The same narrow LF attributes from #350 are included here; real Git autocrlf checkout replay and both focused tests passed on AWS. No production matcher normalization or startup-budget change was made. The existing #305 investigation remains untouched.

Reports were regenerated with explicit latest-track selection. A fresh checkout removed stale workspace-plan references to untracked upstream lockfiles; the resulting workspace JSON matches #350. Every implementation, integration, and metadata diff passed independent Codex review through P2. Final exact-head hosted CI must pass before landing.

Fixture Impact

  • Updates compatible fixture npm pins and locks
  • Updates submodule pins or the promoted host
  • Publishes packages, tags, or a project version

Incorporates #318, #302, #297 and the compatible new Dependabot updates #328–#346, #348, and #349. Major consumer migrations remain separate; no external plugin source is rewritten.

Final Landing

Required Check run 35729654941 passed at e1c07a604d2daa85ce3444dd07eef9c956676418: Linux, macOS, Windows, container, security, and all 26 isolated changed-fixture lanes. All three advisory canaries also passed. The complete final diff passed a two-part independent P2 review with no actionable findings.

Merged normally as 61be4f5165d2ed2dcb65fc5821380ab57aaf86bc; the entire merged tree is identical to the tested head. #350 is superseded by this complete landing. Its failed standalone Windows evidence is preserved, and no startup/cleanup repair or #305 resolution is claimed. No release, tag, or project-version bump was performed.

@clawsweeper

clawsweeper Bot commented Aug 31, 2026 •

Copy link
Copy Markdown
Contributor

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

ClawSweeper review complete

ClawSweeper finished reviewing this revision. The review result is being finalized.

View the workflow run.

@clawsweeper clawsweeper Bot added merge-risk: 🚨 automation 🚨 Merging this PR could break CI, automerge, proof capture, label sync, or automation. merge-risk: 🚨 compatibility 🚨 Merging this PR could break existing users, config, migrations, defaults, or upgrades. merge-risk: 🚨 security-boundary 🚨 Merging this PR could weaken sandboxing, authorization, credentials, or sensitive data. P2 Normal priority bug or improvement with limited blast radius. proof: sufficient Contributor real behavior proof is sufficient. rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. status: ⏳ waiting on author ClawSweeper has contributor-facing work open and is waiting for author action. labels Aug 31, 2026
@clawsweeper

clawsweeper Bot commented Aug 31, 2026 •

Copy link
Copy Markdown
Contributor

Codex review: needs maintainer review before merge. Reviewed September 22, 2026, 8:56 AM ET / 12:56 UTC (Revision 6).

ClawSweeper review

What this changes

Refreshes plugin fixture packages and locks, repairs telemetry dependencies, adopts Inspector capture fixes, and updates diagnostic expectations, probe prerequisites, and hydration tooling.

Merge readiness

✅ Ready for maintainer review

The refresh remains useful and is not implemented on main. Supplied runtime evidence supports the changes, prior entrypoint concerns are addressed, and no actionable patch defect was found.

Priority: P2
Reviewed head: e1c07a604d2daa85ce3444dd07eef9c956676418

Review scores

Measure Result What it means
Overall readiness 🐚 platinum hermit (4/6) A useful, bounded maintenance change with convincing runtime evidence and no actionable correctness findings.
Proof confidence 🦞 diamond lobster (5/6) Sufficient (terminal): Supplied clean-install Jaeger output and AWS isolated fixture runs exercise the changed dependency and capture paths; linked source/packed Inspector output and real autocrlf replay cover the runtime adoption and fixture-byte correction. Blocked probes remain explicit coverage gaps. No stored-data contract changes.
Patch quality 🐚 platinum hermit (4/6) No actionable review findings were identified.

Verification

Check Result Evidence
Real behavior Verified Sufficient (terminal): Supplied clean-install Jaeger output and AWS isolated fixture runs exercise the changed dependency and capture paths; linked source/packed Inspector output and real autocrlf replay cover the runtime adoption and fixture-byte correction. Blocked probes remain explicit coverage gaps. No stored-data contract changes.
Evidence reviewed 11 items Repository policy: Read the complete root AGENTS.md and searched for nested policies and maintainer notes; none were found for the changed paths. Applied fixture-driven configuration, external-source ownership, credential-free checks, and synchronized Inspector expectations.
Continuing necessity and release boundary: Live main remains 88ab1f2. Its Inspector source pin and fixture packages precede this refresh. Latest release v0.2.1 selects Inspector 0.3.16; neither that release nor current main contains the complete requested change.
Introduced scope and review continuity: Inspected the pinned main-to-head changes and original package-lock data. The verified test merge records the pinned main followed by this exact head. GitHub's head-commit patch confirms that the latest revision adds two LF attributes and corrects generated metadata; the preceding completed review recorded no findings or rank-up moves.
Findings None None.
Security None None.

How this fits together

Crabpot checks OpenClaw plugin compatibility using configured fixtures and a pinned Inspector. Package acquisition, registration capture, and isolated probes produce reports and CI policy results.

flowchart TD
  A[Plugin packages and locks] --> C[Pinned Inspector]
  B[Fixture configuration] --> C
  C --> D[Registration capture]
  D --> E[Prerequisite checks and probes]
  E --> F[Pass, failure, or coverage gap]
  F --> G[Reports and CI policy]
Loading

Before merge

None.

Agent review details

Security

None.

Review metrics

Metric Value Why it matters
Fixture refresh scope 24 package pins updated; 26 locks changed The refresh spans compatibility fixtures while preserving the host, Discord fallback, and external Git pins.
Production and test delta Production script +1/-1; test code +56/-18; captured fixtures +80 Runtime adoption remains a single source-pin change, with focused diagnostic and prerequisite regression coverage.

Technical review

Best possible solution:

Keep dependency refreshes reproducible through fixture locks and upstream Inspector fixes, while preserving host ownership and explicit compatibility gaps.

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

Not applicable to the dependency refresh as a whole; the capture repairs have specific before/after diagnostic and runtime evidence, which this read-only review did not rerun.

Is this the best way to solve the issue?

Yes: the branch uses existing fixture, prerequisite, and failure-classification mechanisms and adopts focused upstream fixes rather than adding parallel runtime logic.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning medium; reviewed against 88ab1f2a9af0.

Labels

Label changes:

No label changes.

Label justifications:

  • P2: Repairs bounded fixture compatibility and dependency-audit problems without changing the promoted host or deployed user workflows.
  • rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🦞 diamond lobster and patch quality is 🐚 platinum hermit.
  • status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (terminal): Supplied clean-install Jaeger output and AWS isolated fixture runs exercise the changed dependency and capture paths; linked source/packed Inspector output and real autocrlf replay cover the runtime adoption and fixture-byte correction. Blocked probes remain explicit coverage gaps. No stored-data contract changes.
  • proof: sufficient: Contributor real behavior proof is sufficient. Supplied clean-install Jaeger output and AWS isolated fixture runs exercise the changed dependency and capture paths; linked source/packed Inspector output and real autocrlf replay cover the runtime adoption and fixture-byte correction. Blocked probes remain explicit coverage gaps. No stored-data contract changes.

Evidence

What I checked:

  • Repository policy: Read the complete root AGENTS.md and searched for nested policies and maintainer notes; none were found for the changed paths. Applied fixture-driven configuration, external-source ownership, credential-free checks, and synchronized Inspector expectations. (AGENTS.md:1, e1c07a604d2d)
  • Continuing necessity and release boundary: Live main remains 88ab1f2. Its Inspector source pin and fixture packages precede this refresh. Latest release v0.2.1 selects Inspector 0.3.16; neither that release nor current main contains the complete requested change. (scripts/plugin-inspector-source.mjs:7, 88ab1f2a9af0)
  • Introduced scope and review continuity: Inspected the pinned main-to-head changes and original package-lock data. The verified test merge records the pinned main followed by this exact head. GitHub's head-commit patch confirms that the latest revision adds two LF attributes and corrects generated metadata; the preceding completed review recorded no findings or rank-up moves. (.gitattributes:1, e1c07a604d2d)
  • Direct Inspector dependency: The source resolver directly imports and prepares openclaw/plugin-inspector. The branch advances its exact source revision while retaining the published 0.3.26 package pin, explicitly documenting the unreleased source additions. Both generated workspace reports equal main after replacing only the Inspector SHA. (scripts/plugin-inspector-source.mjs:7, e1c07a604d2d)
  • Verified upstream repairs: The adopted comparison contains two focused repairs: discovery within packaged .setup directories and preservation of mocked Zod enum options. GitHub verifies both corresponding PRs as merged: fix: discover SDK imports in packaged setup modules plugin-inspector#89 and fix: preserve mocked Zod enum options plugin-inspector#90. These additions address packaged entrypoint capture without adding dependencies or installation hooks. (src/sdk-mock.js:405, 1d27e6906e83)
  • Narrow failure classification: Replacement patterns require the known HAPI or TencentDB diagnostic envelope. Regression cases reject timeouts, cancellation, truncation, signals, unexpected stdout, altered classes, and unrelated errors. The existing classifier preserves rawExitCode and blockedBy; fixture-scoped prerequisite lookup preserves explicit empty-map overrides. (test/fixture-failure-contracts.test.mjs:15, e1c07a604d2d)

Likely related people:

  • unknown: The claimed source-line change could not be verified from bounded local history. (role: source history unknown; confidence: low)
  • unknown: The claimed source-line change could not be verified from bounded local history. (role: source history unknown; confidence: low)

Rating scale

Score Internal tier Crab rank Meaning
6/6 S 🦀 challenger crab Exceptional readiness
5/6 A 🦞 diamond lobster Very strong readiness
4/6 B 🐚 platinum hermit Good normal PR; ordinary maintainer review
3/6 C 🦐 gold shrimp Useful, but confidence is limited
2/6 D 🦪 silver shellfish Proof or implementation needs work
1/6 F 🧂 unranked krab Not merge-ready
N/A NA 🌊 off-meta tidepool Rating does not apply

Overall follows the weaker of proof and patch quality.
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.

Workflow

  • 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.

History

Review history (5 earlier review cycles)
  • reviewed 2026-08-31T07:53:45.420Z sha 372f721 :: found issues before merge. :: [P1] Wait for a Teams-compatible inspector release
  • reviewed 2026-08-31T08:11:27.734Z sha e8e74d6 :: found issues before merge. :: [P1] Pin an inspector release that recognizes the Teams CJS entry
  • reviewed 2026-08-31T08:23:02.308Z sha e8e74d6 :: found issues before merge. :: [P1] Pin an inspector release that handles the refreshed entrypoints
  • reviewed 2026-09-22T10:46:29.836Z sha fefde8a :: needs maintainer review before merge. :: none
  • reviewed 2026-09-22T12:39:48.673Z sha 2c4e115 :: needs maintainer review before merge. :: none

@clawsweeper clawsweeper Bot removed the merge-risk: 🚨 security-boundary 🚨 Merging this PR could weaken sandboxing, authorization, credentials, or sensitive data. label Aug 31, 2026
@clawsweeper clawsweeper Bot mentioned this pull request Sep 9, 2026
8 tasks
Refresh 25 fixture pins and compatible locks, update pnpm tooling, and regenerate current-track reports. Keep OpenClaw host ownership in the workspace plan and remove ineffective HAPI overrides while updating mutable Hono.

Patch Jaeger within its stable major while retaining TencentDB 1.0.1 and SDK 0.218.0. The old SDK's configured propagation path passes a real synthetic trace round trip, and the unchanged lock-security gate is clear.

Validated the full 19-step pinned-host suite in AWS, including 418 tests with 12 expected platform skips, strict fixture smoke, published Inspector smoke, zero breakages, and zero policy failures. Independent three-pass review is scoped-clean through P2.

Related: #318, #302, #297.
@steipete
steipete force-pushed the chore/deps-refresh-20260830 branch from e8e74d6 to fefde8a Compare September 22, 2026 10:42
@steipete steipete changed the title chore(deps): refresh fixtures and enforce inspector smoke checks chore(deps): refresh stable fixtures and repair telemetry audit Sep 22, 2026
@steipete
steipete marked this pull request as ready for review September 22, 2026 10:42
@steipete
steipete requested a review from a team as a code owner September 22, 2026 10:42
@steipete
steipete requested a review from vincentkoc September 22, 2026 10:42
@clawsweeper clawsweeper Bot added rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. and removed status: ⏳ waiting on author ClawSweeper has contributor-facing work open and is waiting for author action. merge-risk: 🚨 automation 🚨 Merging this PR could break CI, automerge, proof capture, label sync, or automation. rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. merge-risk: 🚨 compatibility 🚨 Merging this PR could break existing users, config, migrations, defaults, or upgrades. labels Sep 22, 2026
This was referenced Oct 11, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

P2 Normal priority bug or improvement with limited blast radius. proof: sufficient Contributor real behavior proof is sufficient. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant