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 and safety-conscious, but its guidance is abstract: no concrete refactor technique, command, or example is given, and the single referenced file for detailed patterns is missing from the bundle. The out-of-context "Behavior preservation and evidence" caveats and the verbatim-duplicated intro add tokens without adding actionable instruction.
Suggestions
Inline one or two concrete, executable patterns (e.g., a specific duplication-detection approach or the exact test/regression command to run) since the referenced `resources/implementation-playbook.md` does not exist in the bundle — currently no concrete technique is available anywhere.
Fix or remove the dangling reference: both the Instructions and Resources sections point to `resources/implementation-playbook.md`, but no such file (or references/ directory) is present.
Delete the verbatim repeat of the frontmatter description at the top of the body and trim the "Behavior preservation and evidence" section, which discusses missing examples (EmailService, Logger, CodeQL, discount logic) that appear nowhere in the skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body opens by duplicating the frontmatter description verbatim ("You are a code refactoring expert specializing in clean code principles, SOLID design patterns...") and the Context section ("The user needs help refactoring code to make it cleaner, more maintainable...") restates the same purpose, while "Behavior preservation and evidence" inlines dense caveats about material ("EmailService and Logger are missing", "generic execute($QUERY)") that exists nowhere in the skill. The remaining sections are lean, so it is mostly efficient but has several padded or out-of-context passages that could be trimmed — matching anchor 3 rather than 2 because the bulk of the body is not explanatory filler. | 3 / 5 |
Actionability | Instructions are high-level directives only — "Identify high-impact refactor candidates and risks", "Break work into small, testable steps", "Validate with tests and targeted regression checks" — with no concrete technique, command, or worked example of a refactor, which matches anchor 2's "high-level hints but missing the specific steps to execute". This is not a 3 because there is no pseudocode or partially complete concrete guidance either, and the promised detailed patterns live in `resources/implementation-playbook.md`, a file that does not exist in the bundle. | 2 / 5 |
Workflow Clarity | A rough sequence exists (identify candidates → break into steps → apply → validate with tests → produce a plan per the Output Format section), but validation checkpoints are implicit — no test command, no "only proceed when tests pass" gate, and no fix-and-retry loop for a skill that makes batch code changes. This matches anchor 3 ("steps listed but validation gaps; sequence present but checkpoints missing or implicit") and falls short of anchor 4's "most checkpoints present". | 3 / 5 |
Progressive Disclosure | Section structure is good (Use/Do-not-use, Instructions, Safety, Output Format, Resources) and the external reference is clearly signaled in both Instructions ("If detailed patterns are required, open `resources/implementation-playbook.md`") and a dedicated Resources section — but that referenced file does not exist in the bundle (no references/, scripts/, assets/, or resources/ directory), and the "Behavior preservation and evidence" section inlines commentary that clearly belongs in the missing playbook. This sits between anchors 2 and 3: better organized than "minimal structure", but the dangling reference and misplaced inline content keep it below 4. | 3 / 5 |
Total | 11 / 20 Passed |