| name | os-self-test |
| description | Verify the AI Dev Operating System is in a coherent state inside a project. Detects missing canonical files, broken cross-references, version drift, gitignore gaps, stale paths after a migration, and orphaned artifacts. Run after major edits to the OS or before opening a new sprint. |
OS Self-Test
When to run this skill
- After the user manually edited the OS (rules, agents, skills, commands).
- Before opening a new sprint.
- Before tagging a release.
- When something feels off ("Claude tá agindo estranho").
Checks
1. Canonical structure present
2. Required artifacts (if a project is in progress)
3. Version consistency
4. Gitignore baseline
5. Cross-reference integrity
6. Skill / agent header sanity
Output format
🩺 OS Self-Test — [date]
Canonical structure: ✅
Required artifacts: ⚠️ 3 of 6 present (project may be early — OK if pre-coding)
Version consistency: ✅
Gitignore baseline: ✅
Cross-reference scan: ❌ 2 issues found:
• docs/foo.md links to deleted skills/secrets-scan/
• README badge points to RELEASE-NOTES-v0.1.0.md (deleted)
Skill / agent headers: ✅
Action needed: fix the 2 cross-reference issues above before next sprint.
Implementation note
When asked to run, Claude should:
- Glob the canonical paths.
- Grep for deleted-path references using known migration mappings.
- Read the README badge URL and verify the file exists.
- Read CHANGELOG top entry and compare versions.
- Produce the report above with specific line numbers for any failures.