Content
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A well-organized TDD reference with clear rules and anti-patterns, but it is held back by restated known concepts, implicit validation checkpoints, and references to missing bundle files (tests.md, mocking.md).
Suggestions
Add the referenced bundle files tests.md and mocking.md (or remove the links) so the signaled navigation actually resolves.
Make the red→green loop an explicit numbered checklist with a validation checkpoint, e.g. 'run the test, confirm it fails (red), then write minimal code'.
Tighten the 'What a good test is' section to skill-specific guidance, cutting general behavior-vs-implementation explanation Claude already knows.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is reasonably lean and well-sectioned, but the 'What a good test is' prose restates behavior-over-implementation concepts Claude already knows and could be tightened without losing value. | 2 / 3 |
Actionability | Concrete rules and a worked anti-pattern example ('expect(add(a, b)).toBe(a + b)') plus a specific question to ask users, but as a methodology skill it offers direction rather than copy-paste executable guidance. | 2 / 3 |
Workflow Clarity | The red→green cycle is sequenced ('Red before green', 'One slice at a time'), but the validation checkpoints are implicit — it does not spell out 'run the test and confirm it fails before implementing'. | 2 / 3 |
Progressive Disclosure | References to tests.md and mocking.md are well-signaled and one level deep, but those bundle files do not exist, so the navigation the body promises is broken and cannot reach score 3. | 2 / 3 |
Total | 8 / 12 Passed |