← ClaudeAtlas

health-checklisted

Run a comprehensive codebase health check — tests, linting, coverage, complexity, and dependency audit. Compare with previous runs to detect regressions.
fjpulidop/specrails-core · ★ 9 · AI & Automation · score 76
Install: claude install-skill fjpulidop/specrails-core
Run a full health check for **this project (read from CLAUDE.md or package.json)**: detect available tools, execute each quality check, compare results against the previous run, detect regressions, compute a health grade, and store a snapshot for future comparison. **Input:** $ARGUMENTS — optional flags: - `--since <date>` — use the report from this date (ISO format: YYYY-MM-DD) as the comparison baseline instead of the most recent - `--only <checks>` — comma-separated subset to run. Valid values: `tests`, `coverage`, `lint`, `complexity`, `deps`, `perf`, `static` - `--save` — always save the snapshot even when `--only` is used (default: skip save for partial runs) --- ## Phase 0: Argument Parsing Parse `$ARGUMENTS` to set runtime variables. **Variables to set:** - `COMPARE_DATE` — string (ISO date) or empty string. Default: `""` (use most recent report). - `CHECKS_FILTER` — array of check names or the string `"all"`. Default: `"all"`. - `SAVE_SNAPSHOT` — boolean. Default: `true` when `CHECKS_FILTER="all"`, `false` for partial runs unless `--save` is present. **Parsing rules:** 1. Scan `$ARGUMENTS` for `--since <date>`. If found, set `COMPARE_DATE=<date>`. Strip from arguments. 2. Scan for `--only <checks>`. If found: - Split `<checks>` on commas to produce an array. - Validate each entry against the allowed set: `tests`, `coverage`, `lint`, `complexity`, `deps`, `perf`, `static`. - If any unknown value is found: print `Error: unknown check "<value>". Valid c