fix(test): stabilize tooling guard probes#95114
Conversation
|
Codex review: needs maintainer review before merge. Reviewed June 19, 2026, 6:53 PM ET / 22:53 UTC. Summary PR surface: Tests +26, Other +6. Total +32 across 3 files. Reproducibility: yes. from source inspection. Current main streams a Kitchen Sink probe response without releasing the reader, and the guard workflow test still searches only the wrapper script for a memberships endpoint now owned by the shared helper; I did not run Vitest in this read-only review. Review metrics: none identified. Merge readiness Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch. Rank-up moves:
Risk before merge
Maintainer options:
Next step before merge
Security Review detailsBest possible solution: Land the narrow script/test fix after the PR is marked ready and exact-head required checks complete successfully. Do we have a high-confidence way to reproduce the issue? Yes from source inspection. Current main streams a Kitchen Sink probe response without releasing the reader, and the guard workflow test still searches only the wrapper script for a memberships endpoint now owned by the shared helper; I did not run Vitest in this read-only review. Is this the best way to solve the issue? Yes. The PR uses the narrow local finally-release fix that matches the sibling helper invariant, and the guard test update follows the current split between security-sensitive-guard.mjs and guard-shared.mjs without changing runtime behavior. AGENTS.md: found and applied where relevant. Codex review notes: model internal, reasoning high; reviewed against 7ee1dafd4fd9. Label changesLabel justifications:
Evidence reviewedPR surface: Tests +26, Other +6. Total +32 across 3 files. View PR surface stats
What I checked:
Likely related people:
What the crustacean ranks mean
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
|
|
Land-ready proof for
The earlier exact-head gate Local targeted Vitest was not run in this Codex worktree because |
* fix(test): release kitchen sink probe readers * test(github): follow shared guard membership helper
* fix(test): release kitchen sink probe readers * test(github): follow shared guard membership helper
Summary
response.text()Verification
node --check scripts/e2e/kitchen-sink-rpc-walk.mjsnode --check --experimental-strip-types test/scripts/kitchen-sink-rpc-walk.test.tsnode --check --experimental-strip-types test/scripts/security-sensitive-guard-workflow.test.tsgit diff --checknode scripts/run-vitest.mjs test/scripts/kitchen-sink-rpc-walk.test.tswas not run locally because this Codex worktree has nonode_modulesand exits withOPENCLAW_MISSING_VITEST; dependencies were intentionally not installed in the Codex worktree.The first exact-head CI release gate on
2823bd6f8dafailed inchecks-node-core-toolingon unrelated stale guard workflow coverage (test/scripts/security-sensitive-guard-workflow.test.tsexpected/memberships/in the security-sensitive guard script after the helper had moved toscripts/github/guard-shared.mjs). This PR now includes that narrow test fix.