Content
71%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, well-structured skill body with an exemplary progressive-disclosure split — a concise overview that defers examples to a real, clearly labeled reference file. Its weaknesses are a fully duplicated check list (Check vs. Code Review) and an Explain section that tells Claude things it already knows.
Suggestions
Merge the "Check" and "Code Review" sections — they list the same five checks (heading hierarchy, semantic elements, JS-free content, JSON-LD, hidden content) — and keep one as the audit list.
Trim or repurpose the "Explain" section: that LLMs extract text from HTML and prefer semantic markup is background Claude already has; keep only the checklist-specific rationale.
Add one minimal inline example (e.g., a two-line JSON-LD FAQPage snippet) so the body's most important fix is executable without opening the reference.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly lean, but "Check" and "Code Review" repeat nearly the same five items (heading hierarchy, semantic elements, content visible without JS, JSON-LD, hidden modal/tab content), and "Explain" restates concepts Claude already knows ("Large language models and answer engines process web content by extracting text from HTML"). It could be meaningfully tightened, so it sits below the efficient anchor 4. | 3 / 5 |
Actionability | Guidance is concrete and specific — exact tags to check ("<h1>–<h6>, <p>, <ul>, <ol>, <table>"), exact schema types ("JSON-LD schema (Article, FAQPage, HowTo)"), and specific fixes ("Replace JavaScript-rendered content with server-side rendered HTML"). It stops short of fully executable copy-paste material in the body since all code examples are deferred to references/rule.md, so it does not reach anchor 5. | 4 / 5 |
Workflow Clarity | The Check → Fix → Explain → Code Review flow gives a clear sequence for a simple, single-purpose audit skill, and no validation cap applies since the task is non-destructive. However, the near-duplicate "Check" and "Code Review" lists leave mild ambiguity about which list actually drives the audit, keeping it below the fully unambiguous anchor 5. | 4 / 5 |
Progressive Disclosure | The 45-line body is a well-organized overview with a clearly signaled, one-level-deep reference: "For full implementation details, code examples, and framework-specific guidance, see `references/rule.md`". The referenced file exists and holds the good/bad HTML examples and detailed guidance, exactly the split the top anchor rewards. | 5 / 5 |
Total | 16 / 20 Passed |