Content
100%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.
A lean, dense, fully actionable review skill that defines an exact output format with concrete examples and clear termination criteria. It assumes Claude's intelligence, stays within its niche, and is well-organized for a simple single-purpose skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Roughly 45 lines with no concept over-explanation; it assumes Claude's competence, defining only the output format and tag semantics, and every section earns its place. | 5 / 5 |
Actionability | Gives a concrete output spec ('L<line>: <tag> <what>. <replacement>.'), five precisely defined tags, and several copy-ready ✅ examples covering the common cases; Claude knows exactly what to produce. | 5 / 5 |
Workflow Clarity | Single-purpose, unambiguous action (review diff, emit tagged one-liners, terminate with 'net: -<N> lines possible' or 'Lean already. Ship.'); the destructive-cap does not apply since it explicitly lists rather than applies fixes. | 5 / 5 |
Progressive Disclosure | Under 50 lines with no need for external references, organized into clearly labeled sections (Format, Examples, Scoring, Boundaries); no bundle files exist to evaluate, and none are needed. | 5 / 5 |
Total | 20 / 20 Passed |