Use when the user reports misbehaving software to investigate — a bug, crash, wrong output, flaky behavior, or '為什麼壞掉/不會動/查一下' — and the root cause is not yet established. Enforces building a red-capable reproduction command BEFORE any hypothesis is allowed, then 3-5 ranked falsifiable hypotheses before testing any single one. NOT for conceptual questions, code reading requests, or feature work.
Install
npx skills add KerberosClaw/kc_ai_skills --skill diagnosenpx skills add KerberosClaw/kc_ai_skills --skill diagnose -a claude-code -gnpx skills add KerberosClaw/kc_ai_skills --skill diagnose -a codex -ggemini skills install https://github.com/KerberosClaw/kc_ai_skills.git --path diagnosegh skill install KerberosClaw/kc_ai_skills diagnosegit clone --depth 1 https://github.com/KerberosClaw/kc_ai_skills /tmp/KerberosClaw__kc_ai_skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/KerberosClaw__kc_ai_skills/diagnose ~/.claude/skills/diagnose # Claude Code
cp -R /tmp/KerberosClaw__kc_ai_skills/diagnose ~/.agents/skills/diagnose # Codex, Cursor, Gemini CLI, Copilot, OpenCode