Reviewer Skill Readiness Audit

Current quality snapshot

Why this matters for agent skills

Agent skills work best when prompts include clear trigger conditions, deterministic checklist-style actions, and an expected response format. The current corpus is strong on examples, but weaker on explicit execution contracts (what an agent should output) and standardized structure.

  1. Standardize a skill wrapper section during export
  2. Introduce an optional strict mode linter
  3. Segment by prompt length
  4. Leverage JSON datasets for synthesis
  5. Add confidence metadata

JSON dataset signal