Content
77%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 gives concrete, copy-paste-ready command guidance with useful troubleshooting notes, but it depends on a search.mjs script that is not actually bundled and carries minor redundancy with the description.
Suggestions
Bundle search.mjs under scripts/ (or another bundle directory) — the body lists and instructs running it, but the file is missing from the skill package, breaking the one reference and making the commands non-executable as shipped.
Trim redundancy with the frontmatter: the opening line and 'Output expectations' restate 'concise summary' and 'full URLs' already in the description; keep only the non-obvious bits such as highlighting source disagreements.
Once the script is bundled, add a one-line pointer to it (e.g., relative path) so the single external reference resolves cleanly.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and free of concept-explaining fluff, but the opening sentence and 'Output expectations' restate 'concise summary' and 'full URLs' already promised in the frontmatter description, so it could be tightened. | 2 / 3 |
Actionability | Provides an exact command, two concrete examples, and a complete flag list with values — copy-paste ready and not pseudocode. | 3 / 3 |
Workflow Clarity | A simple single-action skill with an unambiguous command and a Notes section covering failure modes (module-resolution env vars, OAuth cached-token fallback), which the rubric allows to score 3 for clear simple skills. | 3 / 3 |
Progressive Disclosure | Sections are well organized and the reference to search.mjs is clearly signaled at one level, but the referenced script file is absent from the bundle (no scripts/ directory), so the reference is broken and navigation to the artifact fails. | 2 / 3 |
Total | 10 / 12 Passed |