Agent skill
testing-guidelines
How tests should be structured, named, and prioritized in this codebase.
Stars
163
Forks
31
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/testing-guidelines
SKILL.md
Testing Guidelines
Test Types
-
Unit tests
- Fast, isolated, no network or DB.
- Test one behavior per test.
-
Integration tests
- Exercise real integrations (DB, queue, external service stubs).
- Focus on critical paths and failure modes.
-
End-to-end (E2E) tests
- Cover full user journeys.
- More expensive; keep the set small but meaningful.
Naming & Structure
- Test files:
- Mirror source structure:
src/foo/bar.py→tests/foo/test_bar.py
- Mirror source structure:
- Test names:
- Use descriptive names reflecting behavior, e.g.:
def test_rejects_requests_without_authentication():
- Use descriptive names reflecting behavior, e.g.:
Coverage Expectations
- Core domain logic: ≥ 80% line coverage
- Peripheral or legacy code: best-effort; prioritize stability
- Do not chase metrics blindly; focus on risk and impact.
Fixtures & Data
- Prefer factory functions over static fixtures.
- Make fixtures explicit and readable; avoid hidden magic.
Testing Conventions
- Arrange-Act-Assert structure where possible.
- Avoid brittle tests that depend on:
- Exact error messages
- Implementation details that can change safely
CI Requirements
- All tests must pass before merge.
- E2E tests may run on a separate pipeline if slow; document behavior.
Didn't find tool you were looking for?