| name | test-rico |
| description | Run focused tests for Rico Hunt — backend pytest, frontend build/lint, and vitest. Use when asked to verify a change, run tests, check correctness, or validate a fix without the full smoke suite. |
| triggers | ["user","model"] |
test-rico
Run focused tests for Rico Hunt. This skill is the fast correctness path — it does not start servers or hit live endpoints. It validates the changed area and stops as soon as there is enough evidence.
Quick checks (run these first)
Frontend build (catches TypeScript / import errors fast)
cd apps/web && npm run build
Expected: no errors, route list printed, exit 0.
Frontend lint
cd apps/web && npm run lint
Backend focused tests (core auth/webhook/onboarding isolation)
python -m pytest tests/test_jotform_webhook.py tests/test_jwt_user_isolation.py tests/test_onboarding_state.py -q
These ~94 tests run in ~2s and do not call live APIs or Neon.
Full suites (when explicitly asked)
cd apps/web && npm run test
python -m pytest tests/ -q --tb=short
Known pre-existing failures
tests/test_agent.py and tests/test_agent_runtime.py contain 6 pre-existing failures (TestApplyServiceIndeedMethod, TestDraftAction, TestJobResolution). They are gating failures only if you are changing the agent runtime. Do not silently fix them as part of an unrelated change.
Cost rule
Stop after the focused checks if they pass. Only run the full suites when the change is large, risky, or the user explicitly asks for it.