Content
20%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 is essentially a collection of generic software engineering platitudes with no concrete, actionable content. It tells Claude to do things Claude already knows how to do (apply clean code principles, break work into steps) without providing any novel knowledge, specific patterns, executable examples, or decision frameworks. The referenced implementation playbook could potentially redeem it, but it doesn't exist in the bundle.
Suggestions
Add concrete before/after code examples demonstrating specific refactoring patterns (e.g., Extract Method, Replace Conditional with Polymorphism) rather than abstract instructions.
Remove the 'Context' section and 'when to use/not use' boilerplate — these waste tokens on information Claude can infer. Replace with specific decision criteria or heuristics for prioritizing refactors.
Include a concrete workflow with validation checkpoints, e.g., 'Run existing tests before any change → make one refactor → run tests again → commit if green → proceed to next refactor'.
Either provide the referenced `resources/implementation-playbook.md` bundle file with actual patterns and examples, or inline the most critical patterns directly in the skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is verbose and padded with information Claude already knows. It repeats the system prompt description, includes 'when to use / not use' sections that are meta-guidance rather than actionable instruction, explains obvious context ('The user needs help refactoring code'), and the instructions are generic platitudes ('focus on readability and stability') rather than novel knowledge. | 1 / 3 |
Actionability | The skill provides entirely vague, abstract guidance with no concrete code examples, specific commands, or executable steps. Instructions like 'Identify high-impact refactor candidates' and 'Apply changes with a focus on readability' describe rather than instruct. There are no concrete refactoring patterns, no before/after code examples, and no specific techniques demonstrated. | 1 / 3 |
Workflow Clarity | There is a rough sequence implied (identify → break into steps → apply → validate), and the output format provides some structure. However, the steps are vague with no explicit validation checkpoints, no feedback loops for error recovery, and no concrete criteria for when to proceed or stop. The mention of 'validate with tests' is implicit rather than a clear checkpoint. | 2 / 3 |
Progressive Disclosure | The skill references `resources/implementation-playbook.md` for detailed patterns, which is a reasonable one-level-deep reference. However, no bundle files are provided, so the referenced file doesn't actually exist, and the main content itself lacks enough substance to serve as a useful overview — it's mostly boilerplate rather than a meaningful entry point. | 2 / 3 |
Total | 6 / 12 Passed |