Comprehensive project health diagnostic — collects signals from CI tools, existing reports, deferred issues, and code markers into a prioritized finding report
Perform a comprehensive project health check by collecting signals from multiple sources, aggregating findings into a unified schema, and producing a prioritized report.
This is a read-only diagnostic skill — it does not modify any code. Use /fix-scrub to remediate findings.
$ARGUMENTS - Optional flags:
--source <list> (comma-separated sources; default: all available)--severity <level> (minimum severity: critical, high, medium, low, info; default: low)--project-dir <path> (directory with pyproject.toml; default: auto-detect)--out-dir <path> (default: docs/bug-scrub)--format <md|json|both> (default: both)Valid sources: pytest, ruff, mypy, openspec, architecture, security, deferred, markers
Scripts live in <agent-skills-dir>/bug-scrub/scripts/. Each agent runtime substitutes <agent-skills-dir> with its config directory:
.claude/skills.codex/skills.gemini/skillsIf scripts are missing, run skills/install.sh to sync them from the canonical skills/ source.
openspec sourcepython3 <agent-skills-dir>/bug-scrub/scripts/main.py \
--source <sources-or-omit-for-all> \
--severity <level> \
--project-dir <path> \
--out-dir docs/bug-scrub \
--format both
The orchestrator produces:
docs/bug-scrub/bug-scrub-report.md — human-readable prioritized reportdocs/bug-scrub/bug-scrub-report.json — machine-readable for /fix-scrubSignal Sources:
Severity Levels (descending): critical > high > medium > low > info
Staleness Warnings: Reports older than 7 days trigger a refresh recommendation.
/fix-scrub --tier auto for tool-native auto-fixes (ruff)/fix-scrub for auto + agent-assisted fixes/fix-scrub --dry-run to see what would be fixed/fix-scrub --severity high for high-priority items only/plan-feature proposal for findings that need design decisionspython3 -m pytest <agent-skills-dir>/bug-scrub/tests -q