What problem does it solve?
Manually auditing a growing library of Claude skills and commands for quality, overlap, staleness, and alignment with best practices is extremely time-consuming and prone to oversight, especially for teams managing dozens or hundreds of custom skills.
Core Features & Use Cases
- Dual Audit Modes: Run a Quick Scan to only evaluate recently changed skills (reducing review time to 5-10 minutes) or a Full Stocktake for a complete end-to-end review of all global and project-level skills.
- Holistic Quality Judgment: Uses a standardized checklist and AI evaluation to assess each skill for actionability, scope fit, uniqueness, and technical currency, with clear, actionable verdicts for every skill.
- Use Case: Developer platforms and AI teams can use this skill to eliminate redundant skills, retire outdated workflows, and consolidate overlapping functionality to keep their Claude skill library maintainable and effective.
Quick Start
Use the skill-stocktake skill to run a full quality audit of all your global and project-level Claude skills, or request a quick scan of only recently modified skills to reduce review time.