Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is well-structured and uses progressive disclosure effectively, but it loses points for restating known SEO rationale, omitting executable code and a named scan tool in the main file, and missing an explicit verification step in the batch Fix workflow.
Suggestions
Trim the 'Explain' section and intro of SEO concepts Claude already knows (snippet rewriting, duplicate-content dilution) to improve token efficiency.
Add an explicit verification checkpoint to the Fix workflow, e.g. 'Re-crawl after deploying to confirm no duplicate descriptions remain.'
Name a concrete scan method or tool in the Check step (or surface a one-line command) so the main file is actionable without requiring the reference.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient with short sections, but the intro and 'Explain' section restate SEO concepts Claude already knows (e.g. Google rewriting snippets, pages being 'interchangeable'), so not every token earns its place. | 2 / 3 |
Actionability | Concrete framework pointers (Next.js metadata export, Nuxt useHead, Yoast/RankMath), character limits, and example descriptions are present, but no executable code lives in the body and the scan method is unspecified ('Scan the HTML of every page'), leaving key details incomplete here. | 2 / 3 |
Workflow Clarity | A Check-to-Fix sequence with numbered steps exists, but this site-wide batch operation has no explicit verification/re-crawl checkpoint in the body's Fix steps, which caps the score per the batch-operations guideline. | 2 / 3 |
Progressive Disclosure | The body is a concise overview with clearly signaled one-level-deep navigation to references/rule.md (a real, verified file), with content appropriately split rather than monolithic. | 3 / 3 |
Total | 9 / 12 Passed |