Content
71%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A tightly written, directive body with a clear loop structure and genuinely useful anti-pattern guidance. Its one significant defect is that both of its external references (tests.md, mocking.md) are dangling — the promised examples and mocking guidelines simply don't exist in the skill bundle.
Suggestions
Create the referenced files (tests.md with worked good/bad test examples, mocking.md with mocking guidelines) or remove the references from the body so no navigation path dead-ends.
Add an explicit red-phase checkpoint to the loop rules, e.g. 'Run the test and confirm it fails for the expected reason before writing any implementation.'
Trim the near-duplicate statement between the 'What a good test is' and 'Seams' sections (both assert tests go through public interfaces, not internals).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes competence, teaching by example rather than explanation (e.g. the tautological-test tell 'expect(add(a, b)).toBe(a + b)'). Minor redundancy: 'Tests verify behavior through public interfaces, not implementation details' in the first section largely repeats 'Tests live at seams, never against internals' in the seams section. Not 5: a few sentences could be tightened. | 4 / 5 |
Actionability | Concrete directives throughout — 'Write the failing test first, then only enough code to pass it', the exact question to ask ('What's the public interface, and which seams should we test?'), and a diagnostic 'tell' for each anti-pattern. Not 5: the strongest concrete material (worked examples of good tests, mocking guidelines) is delegated to tests.md and mocking.md, which do not exist in the bundle. | 4 / 5 |
Workflow Clarity | The loop is clearly sequenced (agree seams with the user → write failing test → minimal implementation → repeat), with an explicit pre-cycle checkpoint ('No test is written at an unconfirmed seam') and explicit routing of refactoring to the review stage. Not 5: there is no explicit checkpoint to run the test and confirm it is actually red before implementing. | 4 / 5 |
Progressive Disclosure | Sections are well organized and references are clearly signaled ('See [tests.md](tests.md) for examples and [mocking.md](mocking.md) for mocking guidelines'), but neither referenced file exists — there are no references/, scripts/, or assets/ directories in the bundle. Not 4/5: a signposted reference that leads nowhere is a real navigation gap, and this skill clearly needs those external files for its examples. | 3 / 5 |
Total | 15 / 20 Passed |