skill-retro

Topics in the source

• skill-retro — grade the agent setup from real sessions
• Hard rules
• Scripts
• References and assets

Privacy is the contract. Everything runs locally. Transcripts are condensed, secret-redacted, chmod-0600, and never uploaded — the only shareable artifact is the report the user chooses to share. Run from the repo being graded. Every artifact goes to one fresh scratch dir, never into the repo: 1 — Collect. Scans Claude Code project-history JSONL and Codex rollouts, discovers repo skills (.claude/skills, .agents/skills, .codex/skills, plugin layouts), detects skill usage (Skill invocations, slash commands, SKILL.md paths), samples newest-first, and writes redacted transcripts. Read $RUN/inventory.json: if sessionssampled is 0, tell the user there is nothing recent to score (suggest --days 90 or --repo) and stop. skillsfound 0 is fine — the report becomes a case for creating skills. To grade headless runs (for example skill evals saved as stream-json), add --sessions-dir . 2 — Score. python scripts/scoreaggregator.py --inventory "$RUN/inventory.json" -emit-template > "$RUN/sessionscores.json". Read each transcript in $RUN/transcripts/ and judge it against both rubrics — scorers/efficiency.md and scorers/code-quality.md. Fill the template with a label from the rubric's table and a 1–3

This is an excerpt. Refer to the complete source and its current requirements.