Content
42%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 skill body is well-sectioned (use/do-not-use guidance, safety constraints, and an output format are all present) but offers almost no executable substance — instructions are abstract directives, and the one file that would carry the detailed patterns (resources/implementation-playbook.md) is missing from the bundle entirely. It also pads token budget by repeating the frontmatter persona and restating obvious context.
Suggestions
Either add the referenced `resources/implementation-playbook.md` with the detailed refactor patterns, or remove both references to it — a dangling path in Instructions and Resources makes the guidance unactionable.
Add one concrete anchor to the Instructions, e.g. a short example of a prioritized refactor step (before/after snippet) or a specific technique checklist (extract method, inline variable, guard clauses), so the guidance is executable rather than descriptive.
Make validation an explicit gate with a feedback loop ('run tests after each step; if failing, revert that step before proceeding') and delete the duplicated persona paragraph and Context section, which restate the frontmatter description.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and mostly efficient, but it duplicates the frontmatter description verbatim ("You are a code refactoring expert specializing in clean code principles, SOLID design patterns...") and the Context section restates what is already evident from the title and description ("The user needs help refactoring code to make it cleaner, more maintainable..."). This matches anchor 3 ('mostly efficient but includes some unnecessary explanation or could be tightened') rather than anchor 4, since two whole sections add little. | 3 / 5 |
Actionability | The Instructions are high-level directives ("Identify high-impact refactor candidates and risks", "Break work into small, testable steps", "Apply changes with a focus on readability and stability") with no concrete commands, tooling, techniques, or examples of what a refactor step looks like. This fits anchor 2 ('minimal concrete guidance; high-level hints but missing the specific steps to execute'); it is not score 1 because the output format and safety constraints do add some concrete structure, and it is not score 3 because nothing here is executable — even the promised detailed patterns live in a referenced file that does not exist in the bundle. | 2 / 5 |
Workflow Clarity | A rough sequence is present (identify candidates -> break into steps -> apply changes -> "Validate with tests and targeted regression checks"), and validation is at least mentioned, so this clears anchor 2. But there are no explicit checkpoints or feedback loops (validate -> fix -> re-validate) for an operation class the rubric flags as needing them (batch changes to a codebase), leaving implicit what should be an explicit gate, which matches anchor 3. | 3 / 5 |
Progressive Disclosure | Section structure is clean and the reference is clearly signaled in a dedicated Resources section, but the body twice points to `resources/implementation-playbook.md` and no such file (or any bundle directory) exists in the skill, so the promised detailed patterns are unreachable. Per the guideline to score against the actual bundle structure rather than references alone, this dangling reference prevents a 4-5: navigation is advertised but broken. | 3 / 5 |
Total | 11 / 20 Passed |