fix(machine-health): advertise audit RunMode in argument-hint

[kyle-sexton]

No linked issue

Summary

machine-health:audit already accepts weekly|on-demand|first-run and --dry-run, but autocomplete had no argument-hint. This is the leftover concrete fix from #3542 after #3673.

Fix

• Add argument-hint: "[weekly|on-demand|first-run] [--dry-run]" to plugins/machine-health/skills/audit/SKILL.md.
• Bump machine-health to 0.12.9 and record the change.

House-style convention docs and a check-skill.sh argument-hint criterion stay deferred (#3620 post-pilot gate). This PR does not close #3542.

Verification

scripts/affected-tests.sh --run --explain selected no suites: every changed file is a recorded no-suite class.

affected-tests no-suite result

Related

Refs #3542. Follows #3673 (items 2-5) and #3672 (machine-health 0.12.8 reference/ rename).

_{To show artifacts inline, enable in settings.}

  

[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 2026-09-02T16:22:48.691752Z	ecf8122	PR opened

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.

[claude]

Claude finished @kyle-sexton's task in 52s —— View job

---

Code review complete

• Gather context (PR diff, CLAUDE.md, changed files)
• Apply skip gate
• Review changed files against criteria
• Post findings (inline + summary)

Skip gate: change is trivial with no meaningful review surface. Verified anyway before invoking it:

• argument-hint: "[weekly|on-demand|first-run] [--dry-run]" added to plugins/machine-health/skills/audit/SKILL.md accurately reflects the RunMode (weekly/on-demand/first-run) and DryRun parameters documented at SKILL.md L31-34 — no drift between the hint and the actual accepted arguments.
• The bracket/pipe syntax matches this repo's existing argument-hint conventions (e.g. plugins/playwright/skills/playwright/SKILL.md's "[update] [--check|--apply]").
• Version bump (0.12.8 → 0.12.9) in plugin.json and the corresponding CHANGELOG.md entry are consistent with each other and with the changelog's existing format (Keep a Changelog style, no dates used elsewhere either).
• No code/logic paths touched — pure metadata/doc change.

No findings. Nothing to flag.

[claude]

Claude finished @kyle-sexton's task in 35s —— View job

---

Security review complete

• Gather context (PR diff, changed files)
• Invoke /review:security-review per custom instructions
• Post findings

No security-relevant changes. This PR is a metadata-only change:

• plugins/machine-health/skills/audit/SKILL.md#L3 — adds a static argument-hint frontmatter string ("[weekly|on-demand|first-run] [--dry-run]"). It's display-only text shown in CLI autocomplete, not executed, not derived from user input, and doesn't alter the skill's existing argument parsing/validation.
• plugins/machine-health/.claude-plugin/plugin.json#L4 — version bump 0.12.8 → 0.12.9.
• plugins/machine-health/CHANGELOG.md — changelog entry documenting the above.

No code paths, permissions, trust boundaries, or input handling are touched. No findings.

[github-actions]

Claude has reviewed this PR 1 time. The lane skips further automatic reviews after 5; deleting this comment resets the count.

[github-actions]

Last security-reviewed head: ecf81220e2f55c8048ed6ba1686fcc6ea7c2f519. On the next push, the relevance gate compares only the commits since this SHA; delete this comment to force a full re-review.