Content
62%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a well-structured TDD workflow with clear sequencing and good anti-pattern guidance. Its main weaknesses are the lack of concrete, executable code examples showing an actual TDD cycle in practice, and some verbosity in the philosophy section explaining testing concepts Claude already knows. The references to supporting files are well-placed but unverifiable without bundle files.
Suggestions
Add a concrete, executable code example showing one full RED→GREEN cycle (e.g., a Python/JS test that fails, then the minimal implementation that makes it pass) to replace or supplement the abstract test1→impl1 placeholders.
Trim the philosophy section — Claude already understands integration vs unit testing and implementation coupling. Reduce to 2-3 bullet points of project-specific conventions rather than general testing philosophy.
Provide the referenced bundle files (tests.md, mocking.md, refactoring.md) or remove the references if they don't exist, as broken references reduce trust in the skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The philosophy section spends significant tokens explaining what good vs bad tests are — concepts Claude already understands well. The anti-pattern section, while valuable, is somewhat verbose in its explanation. The ASCII diagrams and checklists are efficient, but overall the skill could be tightened by ~30%. | 2 / 3 |
Actionability | The workflow steps are clear but lack concrete, executable code examples. The test/implementation examples use abstract placeholders (test1→impl1) rather than real code showing an actual TDD cycle. There are no language-specific executable examples demonstrating how to write a failing test and then make it pass. | 2 / 3 |
Workflow Clarity | The workflow is clearly sequenced (Planning → Tracer Bullet → Incremental Loop → Refactor) with explicit validation checkpoints (checklists per cycle, 'run tests after each refactor step', 'never refactor while RED'). The feedback loop of RED→GREEN is well-defined with clear rules about when to proceed. | 3 / 3 |
Progressive Disclosure | References to tests.md, mocking.md, refactoring.md, CONTEXT.md, and /codebase-design skill are present and one-level deep, which is good. However, no bundle files were provided, so these references are unverifiable. The philosophy section is fairly long and could potentially be split out, keeping SKILL.md as a leaner overview. | 2 / 3 |
Total | 9 / 12 Passed |