Test Quality Assessment
Evaluate whether tests verify true functionality rather than rubber-stamping code with superficial assertions.
Layers:
- •
rules.md- Quick reference and severity guidance - •
examples.md- Detailed code examples
Quick Reference
| Factor | Question | Severity if problematic |
|---|---|---|
| Substantive assertions | Does the test prove anything meaningful? | Critical |
| True functionality | Is observable behavior tested, not implementation? | Critical |
| Test organization | Are tests in the right module and cohesive? | Improvement |
| Edge case coverage | Are error paths and boundaries tested? | Critical/Improvement |
| Test data variety | Is data varied or repetitive? | Improvement |
| Fixture usage | Are fixtures reducing duplication without tight coupling? | Improvement |
| Mock discipline | Are mocks used only when necessary? | Improvement/Critical |
| Tests run | Do tests actually pass without errors? | Critical |
Quick Heuristic
A test is valuable if:
- •Failure reveals bugs: The test would fail if the implementation was wrong
- •Survives refactoring: The test wouldn't break if implementation details changed
- •Documents intent: Reading the test explains what the code should do
- •Runs independently: The test passes regardless of execution order
For rules: see rules.md
For examples: see examples.md