Content
78%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 tight, actionable review checklist with concrete API-level guidance and a well-organized structure appropriate for a short, bundle-free skill. The main improvements are removing the description-duplicating opening line, fixing the malformed hook names, and adding example patterns or an explicit findings format.
Suggestions
Drop or compress the opening line that restates the frontmatter description, reclaiming tokens without losing information.
Fix the malformed hook references ('use state', 'use effect' should be 'useState', 'useEffect') and add one or two example code smells per checklist item so the reviewer knows exactly what pattern to flag.
Specify the expected output of a review (e.g., findings format for the 'Review record' PR comment) to complete the workflow.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and directive with no explanations of concepts Claude already knows, but the opening line ('Use this skill when reviewing Next.js App Router structures, component boundaries, state management, and React performance') near-duplicates the frontmatter description and could be trimmed. This matches anchor 4 ('efficient; minor instances... that could be trimmed') rather than 5's 'every token earns its place'. | 4 / 5 |
Actionability | Concrete, executable guidance naming exact APIs ('"use client"', 'URLSearchParams and next/navigation controls', 'internal /api/... endpoints') and specific checks ('un-memoized expensive calculations', 'excessive component mount/unmount cycles'). Minor gaps — no example code smells to flag and the oddly spaced '`use state`, `use effect`' muddies the hook names — place it at anchor 4 rather than 5's fully example-covered level. | 4 / 5 |
Workflow Clarity | A clear sequence exists: protocol prerequisites (AGENTS.md, codex review protocol) -> two numbered checklist sections -> post one 'Review record' comment, with an explicit safety constraint ('Do not review opportunistically or mutate files during pure review'). It sits at anchor 4 rather than 5 because there is no explicit output format for findings and no error-recovery guidance; the destructive-operation cap does not apply since this is a read-only review skill. | 4 / 5 |
Progressive Disclosure | The skill is under 50 lines with no bundle files, and the content is well-organized into clearly signaled sections; its external doc references (AGENTS.md, docs/codex-review-protocol.md, pull-request-workflow.md) are one level deep and clearly labeled. Per the simple-skill guideline this earns a 5; nothing that belongs in a separate file is inlined. | 5 / 5 |
Total | 17 / 20 Passed |