Content
60%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 body is well-structured and concise with a reasonable sequenced workflow and verification checkpoint, but its instructions stay abstract without concrete executable guidance and reference a resource file that is not present in the bundle.
Suggestions
Add concrete, executable guidance to the Instructions (e.g., specific refactor patterns, example diff slices, or commands to run tests) rather than only high-level directives.
Either create resources/implementation-playbook.md or remove the dangling references in the Instructions and Resources sections so navigation is not broken.
Delete the duplicate 'You are a code refactoring expert...' opening paragraph and trim the Context section, since the frontmatter description already conveys this.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean with short bullet sections and no padding of concepts Claude already knows; the only trimmable redundancy is the opening paragraph and 'Context' section re-stating the description already in frontmatter. | 4 / 5 |
Actionability | Guidance is high-level and abstract ('Assess code smells', 'Propose a refactor plan with incremental steps') with no concrete code, commands, patterns, or specific steps to execute, fitting minimal concrete guidance. | 2 / 5 |
Workflow Clarity | Instructions give a clear ordered sequence (assess, plan, apply in slices, update tests/verify) with a verification checkpoint and Safety notes on tests passing, though an explicit error-recovery feedback loop is missing. | 4 / 5 |
Progressive Disclosure | The body is well-organized into clear sections and signals a one-level-deep reference ('resources/implementation-playbook.md'), but that referenced file does not exist in the bundle, a navigation gap that prevents a top score. | 4 / 5 |
Total | 14 / 20 Passed |