Content
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, concise skill that efficiently communicates language-agnostic testing principles. Its main weakness is the lack of any concrete examples—even a single illustrative test name or AAA-structured pseudocode snippet would significantly boost actionability. The organization and token efficiency are excellent.
Suggestions
Add one concrete example of a well-named test following the 'should <behavior> when <condition>' pattern to make the naming convention immediately actionable.
Include a brief, language-agnostic AAA example (even pseudocode) showing the blank-line separation between Arrange, Act, and Assert phases.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every line delivers actionable guidance without explaining concepts Claude already knows. No filler, no definitions of testing or mocking—just direct, lean instructions. | 3 / 3 |
Actionability | The guidance is concrete and specific (e.g., 'Assert both the error type and message', 'Mock only what you don't control') but lacks executable code examples. For a language-agnostic instruction skill this is partially justified, but even one concrete example of test naming or AAA structure would improve actionability. | 2 / 3 |
Workflow Clarity | This is a principles/conventions skill rather than a multi-step process, so workflow complexity is low. The single-task guidance is unambiguous: structure, what to test, mocking rules, assertion rules, and coverage philosophy are clearly sequenced and logically grouped. | 3 / 3 |
Progressive Disclosure | For a short, self-contained skill (~30 lines) with no need for external references, the content is well-organized into clearly labeled sections that are easy to scan. No bundle files are needed. | 3 / 3 |
Total | 11 / 12 Passed |