scorecardlisted

# Scorecard ## Core Rule Render facts that the SessionEnd hook already recorded. Never invent metrics — if a field is missing in the log, surface that, do not synthesize. ## Kit Context Before starting this skill, ensure you have completed session boot: 1. Read `CODEBASE_MAP.md` for project understanding 2. Read `CLAUDE.project.md` if it exists for project-specific rules 3. Confirm `.claude/hooks/session-end.sh` is wired and the SessionEnd profile is active — otherwise the log will be empty If these haven't been read in this session, read them now before proceeding. ## When to Use Invoke with `/scorecard` when: - You want a quick view of the last N sessions (default: 7 days, override with `--window 7d` / `--window 30d` / `--window 14d`) - You want to see which blocking hooks fire most often (signal that the prompt rules need tightening) - You want to track quality-gate failure rate as a moving signal - You want to compare sessions before vs. after a kit upgrade Not for: - Process review or sentiment retrospective — that is `/retro` - Outcome timeline ("what shipped, what broke") — that is `/pulse` - One-off code metrics — those belong in `project-health-report` ## Scope Rules - Reads `reports/session-audit.log` and, when present, `.hook-state/agent-invocations.jsonl` (CLA-38 agent telemetry) — no other sources of truth - Parses both v1 (`schema_version` missing or `1`) and v2 records - Filters by window if provided; default to the last 7 days - Never modifies files