eval-report
Generate evaluation report from session metrics, checkpoint evals, and pass@k data
npx skills add majiayu000/claude-skill-registry --skill eval-report --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Eval Report Generate a dashboard of session effectiveness metrics. ## Arguments - `--days N` — look back N days (default: 7) - `--branch <name>` — filter by branch - `--format table|json` — output format (default: table) ## Process ### 1. Read Metrics Read `.claude/agent-memory/session-learnings/metrics.jsonl` > **Guard:** If `metrics.jsonl` does not exist, output "No session metrics recorded yet. Run sessions with `evaluate-session.sh` enabled to populate." and skip to step 2. Calculate: - Total sessions in period - Average commits per session - Average tasks completed per session - Average files changed per session - Error rate (sessions with errors / total sessions) - Average session duration (minutes) ### 2. Read Checkpoint Evals Read `.claude/phases/output/eval-*.json` Calculate: - Phase transition pass rate - Most common failure criteria - Average attempts per transition ### 3. Read pass@k Data Read `.claude/instincts/eval-tracker.jsonl` Calculate: - pass@1: (tasks succeeded on first attempt) / (total tasks) - pass@3: (tasks succeeded within 3 attempts) / (total tasks) - Average attempts to success ### 4. Skill Effectiveness Analysis Read `.claude/instincts/eval-tracker.jso
- Arguments
- Process
- 1. Read Metrics
- 2. Read Checkpoint Evals
- 3. Read pass@k Data
- 4. Skill Effectiveness Analysis
- 5. Output Report
- Data File Locations
What does the eval-report skill do?
Generate evaluation report from session metrics, checkpoint evals, and pass@k data
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill eval-report --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.