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 lean, highly actionable skill body with a concrete measurement command, explicit thresholds, cause-to-fix mapping, and a properly structured one-level reference to references/rule.md. The only notable issues are the triple repetition of the 15MB limit fact and the absence of a re-measurement verification step after fixes.
Suggestions
Consolidate the 15MB-limit explanation into one place (e.g. keep it in Quick Reference and trim its restatement in the intro and 'Explain' sections) to tighten token efficiency.
Add a final verification step to the Fix workflow, e.g. 'Re-run the curl measurement to confirm the page is under 2MB', and provide a command for counting the byte size of __NEXT_DATA__ script blocks since the Code Review section asks for it.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and domain-specific — explicit thresholds (2MB/5MB/500KB), enumerated causes, and concrete fixes — with no explanation of concepts Claude already knows. Not 5 because the ~15MB crawl limit and its consequences are stated three times (intro, Quick Reference, Explain), and the 'Explain' section largely repeats the intro paragraph. | 4 / 5 |
Actionability | Provides a copy-paste curl command to measure raw HTML size, explicit flag thresholds, and five enumerated causes each paired with a specific fix (externalize SVGs, CDN images, gzip/Brotli, minify, reduce getServerSideProps data). Not 5 because the Code Review section instructs counting __NEXT_DATA__ block bytes without giving a command to do so. | 4 / 5 |
Workflow Clarity | Clear Check → diagnose cause → matching Fix sequence with explicit decision checkpoints (2MB investigate, 5MB critical, 500KB per script block). Not 5 because there is no post-fix verification step (re-measure to confirm the page is under target); the non-destructive nature of the task means the validation cap does not apply. | 4 / 5 |
Progressive Disclosure | A well-sectioned ~50-line overview with a single clearly signaled, one-level-deep pointer: 'For full implementation details, code examples, and framework-specific guidance, see references/rule.md' — and that file exists in the bundle with no further nested references. Content is appropriately split and navigation is easy. | 5 / 5 |
Total | 17 / 20 Passed |