Content
73%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.
The body is a well-crafted, opinionated TDD process guide: concise prescriptive rules, a clearly sequenced workflow with built-in red-green validation loops, and useful checklists. Its one serious defect is that all five inline file references are dangling — the referenced files are absent from the bundle, which breaks progressive disclosure and removes the concrete examples the body defers to them.
Suggestions
Add the referenced files (tests.md, mocking.md, deep-modules.md, interface-design.md, refactoring.md) to the bundle — ideally in a references/ directory — or remove/inline the links so no reference is dangling.
Tighten the Philosophy section to its core directives (e.g., 'Tests verify behavior through public interfaces; they must survive refactors') and move the good-test/bad-test explanations into tests.md, where the body already points for examples.
Include one small concrete red-green example inline (a failing test and the minimal code that turns it green), since tests.md is currently missing and the guidance otherwise never shows what a cycle looks like in practice.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is largely efficient — tight directives ("One test at a time", "Never refactor while RED. Get to GREEN first.") and a compact WRONG/RIGHT diagram — fitting the level-4 anchor 'Efficient; minor instances of over-explanation that could be trimmed'. It is not 5 because the Philosophy section spends paragraphs explaining concepts Claude already knows ("They mock internal collaborators, test private methods...") that could be reduced to the directive alone. | 4 / 5 |
Actionability | As an instruction-only skill it gives concrete guidance — numbered workflow stages, a per-cycle checklist ("[ ] Test describes behavior, not implementation"), explicit rules, and a worked loop diagram — matching the level-4 anchor 'Mostly executable guidance... with minor gaps'. It stops short of 5 because the concrete test examples that would make it copy-paste ready are deferred to tests.md, which does not exist in the bundle. | 4 / 5 |
Workflow Clarity | The process is clearly sequenced (Planning → Tracer Bullet → Incremental Loop → Refactor) with a built-in validation feedback loop ("RED: Write next test → fails / GREEN: Minimal code to pass → passes"), error-recovery rules ("Never refactor while RED", "Run tests after each refactor step"), and checklists for planning and each cycle. This matches the level-5 anchor 'Clear sequence with explicit validation steps; feedback loops for error recovery; checklists for complex processes', with no validation gaps to pull it to 4. | 5 / 5 |
Progressive Disclosure | The body links five files — tests.md, mocking.md, deep-modules.md, interface-design.md, refactoring.md — but none exist anywhere in the bundle (no references/, scripts/, or assets/ directories), so every reference is dangling and navigation is broken. This fits the level-2 anchor ('references are buried'/unusable structure); it is not 3 because 'references present but not clearly signaled' understates the problem — the references are clearly signaled yet point at nothing, which is worse than imperfect signaling of files that exist. | 2 / 5 |
Total | 15 / 20 Passed |