Evaluate Skill

Evaluate whether a Claude Code skill improves task outcomes through repeated paired experiments, blind LLM judging, statistical tests, and a self-contained HTML report. Use when the user asks to benchmark, compare, validate, or quantify the effectiveness of a skill. This is a costly explicit-only workflow.

chenwei791129 3ee3e56 5 files · 85.2 KB Updated

File contents

chenwei791129/agent-skills/tree/main/skills/evaluate-skill commit 3ee3e56776

Frequently asked questions

npx skillmds@latest add chenwei791129/evaluate-skill