Content
62%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 TDD workflow itself (Steps 0–8, runner detection, RED/GREEN gates, checkpoint commits, evidence report) is exceptionally clear and actionable. The skill is dragged down by generic testing-tutorial padding and a monolithic structure with no progressive disclosure — the example libraries and reference material should live in reference files, not the main body.
Suggestions
Move the Testing Patterns, Mocking External Services, E2E patterns, and Test File Organization sections into reference files (e.g. references/patterns.md, references/mocking.md) and link them from a short overview, keeping SKILL.md focused on the workflow steps.
Delete the 'Common Testing Mistakes', 'Best Practices', and 'Success Metrics' sections — they restate testing knowledge Claude already has and add no skill-specific value.
Replace stub examples ('// Implementation here', the empty 'handles database errors gracefully' test) with complete, runnable ones, or explicitly mark them as templates.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Multiple full sections restate testing knowledge Claude already has: a basic Button unit-test pattern, 'Common Testing Mistakes' (test user-visible behavior, avoid brittle selectors, keep tests independent), a 10-item generic 'Best Practices' list (Arrange-Act-Assert, descriptive test names), and a 'Success Metrics' section that restates the coverage requirement. This is noticeably verbose with several padded sections rather than mostly efficient with occasional slack. | 2 / 5 |
Actionability | The runner-detection procedure, command matrix, commit-message formats, and RED/GREEN gate criteria are concrete and executable, and placeholders are explicitly resolved in Step 0. However, several code examples are stubs ('// Implementation here', '// Test implementation', an empty 'handles database errors gracefully' test), so it is mostly rather than fully copy-paste ready. | 4 / 5 |
Workflow Clarity | Steps 0–8 are explicitly sequenced with hard validation gates: the RED state is defined with runtime and compile-time paths and explicit exclusion of unrelated failures, GREEN must be re-verified on the same test target before refactoring, checkpoint commits are tied to each stage, and Step 8 produces an evidence report. This matches the clear-sequence-with-explicit-validation-and-feedback-loops anchor. | 5 / 5 |
Progressive Disclosure | There are no bundle files at all; roughly 230 lines of Testing Patterns, Mocking examples, E2E patterns, and test-file-organization content are inlined in SKILL.md where they belong in separate reference files. Section headers provide reasonable structure, so this sits at 'some structure but content that should be separate is inline' rather than the minimal-structure anchor. | 3 / 5 |
Total | 14 / 20 Passed |