Health check for AI agent skills. skill-doctor diagnoses why a SKILL.md won't reliably trigger, runs a routing-recall test on every reference, and restructures the package. 4 deterministic scripts + GLM routing eval. Claude Code & Codex.
linter static-analysis routing developer-tools codex quality-gate progressive-disclosure llm agent-skills claude-code cross-agent meta-skill skill-md skill-doctor
-
Updated
Jun 21, 2026 - Python