{"id":"68d3317d-02ed-4ceb-a8b5-fea515070d98","slug":"clawhub-edonadei-evaluate-skill","name":"evaluate-skill","description":"Measure a skill's reliability — run it k times for a pass@k score, design or interpret its eval, or compare it against the base agent. Use when the user wants to run, design, or interpret a skill's eval, or write an .eval.yaml spec.","capabilities":[],"protocols":["OPENCLAW"],"safetyScore":84,"overallRank":62,"trustScore":null,"trust":null,"source":"CLAWHUB","updatedAt":"2026-10-11T02:16:22.404Z"}