Content
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is highly actionable with a clear sequenced workflow and an explicit audit-checklist feedback loop, but it is somewhat long for a single-file meta-skill and offers no progressive disclosure beyond well-organized inline sections.
Suggestions
Dedupe the completion-criteria, response-format, capability, and refusal guidance that appears in both the required-sections list and the final audit rubric to tighten conciseness.
Consider moving the worked example or the full audit rubric into a one-level-deep reference file so the SKILL.md overview stays lean and gives progressive disclosure a clear win.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~130-line body is mostly efficient with no concept-explaining fluff, but the nine bad/good example pairs plus the final audit rubric repeat criteria already covered in the required-sections list, so it could be tightened. | 2 / 3 |
Actionability | Concrete and copy-paste-ready throughout: a six-item run contract, nine required sections each with explicit bad/good templates, an audit checklist, and anti-patterns with fixes — all specific actionable guidance for an instruction-only skill. | 3 / 3 |
Workflow Clarity | The 'How to use this skill' section gives a clear four-step sequence (draft contract, write sections, apply audit rubric, reject anti-patterns), and the mandatory audit rubric is an explicit checklist with a feedback loop ('If any item fails, revise the system prompt before writing it'). | 3 / 3 |
Progressive Disclosure | It is a single self-contained file with well-organized section headers and no nested references, but at ~130 lines with all content inline and no bundle files or one-level-deep references, there is no progressive-disclosure structure to push it to 3. | 2 / 3 |
Total | 10 / 12 Passed |