Content
87%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.
The body is lean, well-structured, and actionable with copy-paste examples and a tight parameter reference. Its main gap is the absence of an explicit post-crawl validation checkpoint for what is a batch-fetch operation, which caps workflow clarity.
Suggestions
Add a short verification step after crawling, e.g. 'After crawling, run cache({ query: "<seed topic>" }) to confirm pages landed; if few results, re-run with a broader include_patterns or strategy: sitemap'.
In the Quick Reference, note that strategy: 'sitemap' should be retried with strategy: 'bfs' when the sitemap is missing or sparse, giving an explicit error-recovery branch.
Clarify the relationship between max_pages and max_total_chars so the operator knows which cap will bind first when both are set.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean throughout: a compact parameters table, terse JSON examples, and no padding explaining what crawling, robots.txt, or embeddings are — every section assumes Claude's competence. | 5 / 5 |
Actionability | Quick Reference gives copy-paste-ready JSON for the common cases (sitemap, scoped BFS, URL-map, auth) and the Parameters table pairs each field with a 'When to use' note, covering the typical invocation surface. | 5 / 5 |
Workflow Clarity | The sequence (pick strategy → set scope params → crawl → use cache/find_similar) is present, but this batch operation has no explicit validation/verification checkpoint (e.g. confirm pages cached, verify count vs max_pages), and the rubric caps batch-operation workflow clarity at 3 without one. | 3 / 5 |
Progressive Disclosure | No bundle files exist, and the single-file body is well-organized with clearly labeled sections and one-level-deep 'See Also' links to sibling skills — no nested or buried references, easy to navigate. | 5 / 5 |
Total | 18 / 20 Passed |