Content
32%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 mixes a verbose YAML configuration block with a moderately concrete analysis guide, but lacks a sequenced workflow, executable steps, or any external references. The result is padded and only partially actionable.
Suggestions
Remove or relocate the embedded YAML config block from the body; it is metadata, not instruction, and inflates the token budget without guiding Claude.
Replace the static responsibilities list with a sequenced analysis workflow (scan files -> flag smells by criteria -> compile report) with an explicit validation step.
Add a brief worked example showing the tool calls or commands used to inspect a file, so the guidance is executable rather than purely descriptive.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body opens with a large embedded YAML config block (triggers, capabilities, constraints, hooks, examples) that is config metadata rather than instructions, adding substantial padding before the actual guidance. | 2 / 5 |
Actionability | Concrete criteria are given (e.g. 'Long methods (>50 lines)', an output report template), but there are no executable commands, tool-usage steps, or worked examples showing how to run an analysis. | 3 / 5 |
Workflow Clarity | The numbered 'Key responsibilities' are categories rather than a sequenced process, and there is no analysis workflow with validation checkpoints or error-handling feedback loops. | 2 / 5 |
Progressive Disclosure | All content is inlined in one file with no bundle references; the monolithic YAML config block that opens the body is content that does not belong as instruction and is not well organized for navigation. | 2 / 5 |
Total | 9 / 20 Passed |