test

name	test
description	Create, repair, or run tests for changed behavior.
argument-hint	file, module, feature, or failing test output
license	MIT
metadata	{"category":"quality","tags":["tests","qa","coverage"],"effort":"medium","side_effect":"tests-write","explicit_invocation":false}

Test

Use this skill to add or fix focused tests.

The canonical testing rubric lives in docs/rubrics/testing.md.

Process

1. Detect the test framework and naming conventions.
2. Identify behavior under test.
3. Identify product type and minimum test set from docs/rubrics/testing.md.
4. For changed behavior, write the failing test first and record the red evidence.
5. Add unit tests first, integration tests when boundaries are involved, and E2E/smoke tests for deployable user flows.
6. Include edge cases, negative cases, permission cases, and regression cases for bugs.
7. Add contract tests for API, bot, CLI, webhook, queue, or integration boundaries.
8. Run the smallest relevant test command, then broader tests if risk warrants it.
9. Update TEST_PLAN.md and record green/refactor evidence in QUALITY_GATES.md or .codex-memory/STATE.json.
10. Run or satisfy scripts/validate_plan_quality.py <project> before accepting a behavior-change gate.

Self-validation

Before final output, verify:

• Route, side-effect, and confirmation requirements match metadata.
• Required artifacts or read-only result are explicit.
• Verification, blockers, and next route are stated.

Rules

• Do not rewrite the app to make tests easy.
• Prefer stable behavior assertions over snapshots.
• Mock external services at process boundaries.
• Report commands run and failures that remain.
• Do not mark a gate as passing without a command or explicit manual check.
• Do not count a new behavior test as valid unless it was observed failing before implementation, or the exception is explicit.
• Engineering Discipline v2 blocks behavior changes without TDD red and green evidence.
• For production-facing browser products, ensure /browser-check has Playwright/browser evidence before deploy readiness.