What problem does it solve? Over time, collections of Claude skills and slash commands accumulate stale, overlapping, or low-quality entries, and manually reviewing dozens of skill files is slow and inconsistent. This Skill automates the audit with a fixed quality checklist, holistic AI judgment, and cached results so repeat reviews only re-evaluate what changed. ## Core Features & Use Cases - Two audit modes: Quick Scan re-evaluates only skills changed since the last run using mtime comparison against results.json, while Full Stocktake inventories and evaluates every skill in global and project-level .claude/skills directories. - Structured verdicts: Each skill receives a Keep, Improve, Update, Retire, or Merge verdict with a self-contained reason, evaluated in batches of roughly 20 skills per subagent with resumable progress tracking. - Usage-aware inventory: The scan script enriches each skill with 7-day and 30-day usage counts from an observations JSONL log, plus frontmatter metadata and UTC modification times. - Use Case: Run a Full Stocktake across 80 installed skills, receive a summary table of verdicts, then confirm which overlapping skills to merge and which outdated ones to retire. ## Quick Start Ask the AI to run a full skill stocktake audit of all installed Claude skills and produce a verdict table with consolidation recommendations.
