Content
46%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill covers TDD red-phase test generation comprehensively but is significantly over-explained for Claude's capabilities. Much of the content (edge case categories, anti-patterns, framework conventions, AAA pattern) is knowledge Claude already possesses. The prompt template dominates the file and reads more like a tutorial than a concise skill instruction, and the large amount of inlined reference material would benefit from being split into separate files.
Suggestions
Remove or drastically reduce sections that explain concepts Claude already knows (edge case categories, anti-patterns, quality checklist, basic framework conventions) — keep only project-specific or non-obvious guidance.
Split the detailed prompt template into a separate reference file (e.g., PROMPT_TEMPLATE.md) and keep SKILL.md as a concise overview with the key workflow and tool invocation.
Add explicit feedback loop in the validation workflow: specify what to do when tests fail for wrong reasons (syntax/import errors) vs. right reasons, with concrete commands for each framework.
Provide more concrete guidance on the Task tool invocation with subagent_type='unit-testing::test-automator' — show the exact tool call syntax and expected parameters.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is notably verbose. It explains many concepts Claude already knows well (TDD red phase principles, Arrange-Act-Assert pattern, what edge cases are, anti-patterns like testing implementation vs behavior). The framework patterns section lists basic testing conventions that Claude is deeply familiar with. The edge case categories, quality checklist, and anti-patterns sections are all things Claude inherently understands. | 2 / 5 |
Actionability | The skill provides a concrete TypeScript example with executable code, specifies the subagent tool invocation pattern, and includes a validation workflow. However, the bulk of the 'prompt template' is more of a reference checklist than executable guidance, and the actual mechanism (Task tool with subagent_type) is only mentioned once without detailed usage instructions. | 4 / 5 |
Workflow Clarity | There is a clear sequence in the Instructions section (identify behaviors → generate tests → ensure correct failures → document how to run) and a Validation section with steps. However, the validation steps lack specific commands or concrete checkpoints — 'Run tests - confirm they fail' doesn't specify how to handle failures due to wrong reasons, and there's no explicit feedback loop for fixing setup errors vs. legitimate missing-behavior failures. | 3 / 5 |
Progressive Disclosure | The skill is a monolithic document with no references to external files. The extensive prompt template (framework patterns, edge case categories, quality checklist, anti-patterns, output requirements) could be split into separate reference files. Everything is inlined in a single long file with minimal structural hierarchy beyond headers within the prompt template. | 2 / 5 |
Total | 11 / 20 Passed |