From superpowers-plus
Enforces exhaustive scope enumeration, item-by-item tracking, automated validation, and coverage metrics before claiming any audit, refactoring, or bulk-edit task complete.
How this skill is triggered — by the user, by Claude, or both
Slash command
/superpowers-plus:exhaustive-audit-validationThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
> **Wrong skill?** Quick spot check → `completeness-check`. Pre-commit → `pre-commit-gate`. Repo health → `holistic-repo-verification`.
Wrong skill? Quick spot check →
completeness-check. Pre-commit →pre-commit-gate. Repo health →holistic-repo-verification.Purpose: Prevent "first-pass complete" followed by "found 12 more issues" Root Cause: Agent claimed audit complete without exhaustive validation Incident: 2026-02-28 — "First-pass audit" missed 12 of 27 skills needing fixes
Pattern observed:
Root cause: No systematic validation BEFORE claiming done.
<EXTREMELY_IMPORTANT>
Before claiming ANY audit, bulk-edit, or refactoring task is complete, execute ALL of these steps:
Before starting work, enumerate ALL items that need review:
# Example: Skill trigger audit
find skills -name "skill.md" | wc -l
# Result: 27 skills
# Example: Test file updates
find . -name "*.test.ts" | wc -l
# Result: 43 test files
Record in task list:
"Auditing exactly N items: [list or reference]"
Maintain a checklist showing status of EVERY item:
| # | Item | Status | Notes |
|---|---|---|---|
| 1 | wiki-orchestrator/skill.md | ✅ Fixed | Added triggers |
| 2 | detecting-ai-slop/skill.md | ✅ Fixed | Added triggers |
| 3 | vitest-testing-patterns/skill.md | ✅ Migrated | Moved to overlay repo |
| ... | ... | ... | ... |
Status options:
Execute verification commands that would catch the gaps:
# Example: Check for missing YAML frontmatter
grep -L "^---" skills/*/skill.md skills/*/*/skill.md 2>/dev/null
# Example: Check for skills without "Use when" pattern
grep -L "Use when" skills/*/skill.md skills/*/*/skill.md 2>/dev/null
# Example: Check for skills without YAML triggers field
grep -L "^triggers:" skills/*/skill.md skills/*/*/skill.md 2>/dev/null
# Example: Count skills with proper frontmatter
grep -l "^---" skills/*/skill.md skills/*/*/skill.md 2>/dev/null | wc -l
If validation finds issues → FIX THEM before claiming done
Before marking complete, state:
## Pre-Completion Validation Report
**Scope:** Audited X of Y items (100%)
**Validation checks passed:**
- ✅ All skills have YAML frontmatter
- ✅ All skills have "Use when" pattern
- ✅ All skills have YAML `triggers:` field
**Remaining gaps:** None
**Ready to claim complete:** YES
If gaps remain:
**Remaining gaps:** 3 skills still need triggers
**Ready to claim complete:** NO — fix gaps first
</EXTREMELY_IMPORTANT>
| Task Type | Validation Commands |
|---|---|
| Skill trigger audit | grep -L "^triggers:" skills/*/*/skill.md |
| YAML frontmatter | grep -L "^---" skills/*/*/skill.md |
| Test coverage | npm run test:coverage |
| Lint fixes | npm run lint with zero errors |
| Type errors | npm run typecheck with zero errors |
| Import updates | grep -r "old-import" src/ returns empty |
This skill extends superpowers:verification-before-completion:
When both apply, run this skill FIRST, then verification-before-completion.
If this skill is skipped:
This skill exists because the agent claimed "first-pass complete" on 2026-02-28 while 12 of 27 skills were unfixed.
| Failure | Recovery |
|---|---|
| Checking only the first few instances | EVERY instance must be verified. Use grep counts to confirm. |
| Declaring 'all fixed' without independent count | Count before and after. Numbers must match. |
| Missing instances in non-obvious locations | Check tests, configs, docs, comments — not just source code |
npx claudepluginhub bordenet/superpowers-plus --plugin superpowers-plusGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.