Stats
Actions
Tags
How this skill is triggered — by the user, by Claude, or both
Slash command
/qa-agent:judge-evalsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
The judge exists to prevent fake confidence.
The judge exists to prevent fake confidence.
Required report shape:
{
"happy_path": {"status": "PASS|FAIL", "steps": []},
"adversarial": [
{"rule": "overpayment_protection", "status": "HELD|BREACHED|INCONCLUSIVE", "evidence": {}}
],
"summary": "..."
}
Eval checklist:
happy_path, adversarial, summary.HELD includes evidence of clean rejection and/or valid state after.HELD guardrails.BREACHED finding includes reproduction evidence.Use scripts/eval_report.py report.json to run basic structural evals.
npx claudepluginhub jbcrane13/qa-agent --plugin qa-agentProvides a checklist for code reviews covering functionality, security, performance, maintainability, tests, and quality. Use for pull requests, audits, team standards, and developer training.