Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is well-structured and self-contained with good progressive disclosure, but it loses points on a broken primary code example, duplicated endpoint descriptions, and the absence of any response validation or error-handling guidance. Fixing the malformed Search curl example would be the highest-impact improvement.
Suggestions
Fix the Search curl example so the body is a single valid JSON string inside `-d`, matching the structure of the Fetch example (e.g. wrap q/depth/outputType under a `body` key).
Remove the duplicated endpoint one-liners ("The /search/… endpoint allows you to…") that repeat the Capabilities section verbatim under Usage.
Add brief error/validation guidance, such as checking the response status and what to do when credentials are missing or the API returns an error.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly lean parameter lists and curl examples, but endpoint descriptions are duplicated ("The /search endpoint allows you to retrieve web content" appears in both Capabilities and Usage) and the Use Cases section restates the obvious, so it could be tightened. | 2 / 3 |
Actionability | Concrete curl examples are provided, but the primary Search example is malformed — the request body JSON appears as stray lines outside the `-d` string — so the most important example is not copy-paste executable, matching the 'incomplete guidance' anchor. | 2 / 3 |
Workflow Clarity | The skill is organized as API reference rather than a sequenced workflow, and there are no validation or error-handling checkpoints for API responses (e.g. what to do on auth failure or empty results), leaving the sequence present but checkpoints implicit. | 2 / 3 |
Progressive Disclosure | It is a single self-contained file with no bundle references, organized into clear sections (Setup, Capabilities, Usage, Use Cases, Discover More) and no nested/deep references, which is appropriate navigation for a skill of this size. | 3 / 3 |
Total | 9 / 12 Passed |