[kyle-sexton]

Refs: #8558

Summary

The 28 gaps in the run the issue names (39240292374, shard 4/7) were already fixed by #8411. Re-running the audit on current main (shard 4/7, 132 suites) found 182 gaps in five other suites.

Fix

Each suite declares the files it reads in a # test-scope: header, as #8411 did; no selector logic changes.

• engine.test.sh (178 files): every plugins/*/hooks/* file, which test_babysit_risk.py scans, plus the convention, lib, fixture and workflow-helper files it reads.
• goimports-format.test.sh, markdown-check.test.sh: the plugin's plugin.json.
• setup-apply.test.sh: plugins/verification/lib/parse-concern-value.sh.
• read-lane-broker-check.test.sh: .github/standards/runner-policy/package.json.

Verification

• scripts/selection-audit.sh trace on the five suites: no gap.
• scripts/check-affected.sh --tests: 13 passed, 1 skipped.
• Other shards were not re-traced.

Related

#8411, #8410

🤖 Generated with Claude Code

https://claude.ai/code/session_01Xk7Lz7d2twxvA3LfRxnLkT

Scope note: this covers audit shard 4 of 7 on today's main (182 gaps in 5 suites, 178 in engine.test.sh); the 28 gaps in run 39240292374 were already fixed by #8411. The other six shards were not re-traced, so #8558 stays open until they are.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-20T11:54:01.309690Z	c10b98f	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[github-actions]

claude-security-review reviewed c10b98f5dea3496bfb5b03b6e5d8a961b034cb89: 0 finding(s), 0 as inline comments.

No security issues found. The PR changes only # test-scope: comment headers in five shell test suites, plus four .changes/* fragments (bump: none). It has no executable, workflow, permission, credential or instruction-surface changes. No standing instruction is deleted, narrowed or softened. Reviewed at c10b98f.

[github-actions]

claude-review reviewed c10b98f5dea3496bfb5b03b6e5d8a961b034cb89: 0 finding(s), 0 as inline comments.

No findings. The PR adds # test-scope: headers to five test suites and four bump: none changelog fragments, with no logic changes.

What I checked:

• Every path and glob in the new headers resolves to an existing file. That includes the engine.test.sh additions (the plugins/*/hooks/* glob and the lib, convention, workflow-helper and fixture files), the two plugin.json files, plugins/verification/lib/parse-concern-value.sh and .github/standards/runner-policy/package.json.
• The fragment directories (go-format, markdown-format, source-control, verification) are all listed in scripts/fragment-plugins.txt.
• Not run: I did not run scripts/selection-audit.sh or the suites. The PR's "no gap" and "13 passed, 1 skipped" claims are the author's and unverified by me.
• Scope: the PR says it covers audit shard 4 of 7 only, and the PR text refers to Audit findings: maintenance-audit-test-selection reported findings #8558 staying open for the other shards. The Refs: link keeps it open, so there is no closing-issue criterion to check.