Content
67%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 delivers a well-structured, genuinely additive methodology: a three-layer search strategy with concrete search commands, source-quality criteria, an output template, and anti-patterns. Its main weaknesses are redundancy with the frontmatter description and comment-only pseudo-code blocks instead of executable example commands.
Suggestions
Remove or drastically shrink the 'When to Use This Skill' section — it duplicates the frontmatter description that is always loaded alongside the skill.
Convert the comment-only bash blocks into actual executable example commands (e.g. a real web-search query and a real 'gh search code' invocation) instead of commented illustrations.
Wire the Result Evaluation criteria into the Search Workflow as an explicit checkpoint step (e.g. '4. Evaluate each source against the criteria table before synthesizing').
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient — the grep.app URL pattern, the '(?s)' regex tip, the 1000+ stars heuristic, and the evaluation table all earn their tokens — but the 'When to Use This Skill' section duplicates the frontmatter description and the three bash blocks contain only comments, fitting anchor 3 ('could be tightened') more than anchor 4's minor instances. | 3 / 5 |
Actionability | Concrete, usable guidance throughout: a real grep.app URL pattern, an executable 'gh search code "getServerSession(" --language=TypeScript' command, site: query examples, a complete output-format template, and a worked rate-limiting example. The example queries sit in comment-only blocks rather than ready-to-run commands and 'Synthesize' is underspecified, so it falls short of anchor 5. | 4 / 5 |
Workflow Clarity | The Search Workflow is a clear numbered 4-step sequence (docs → GitHub → web → synthesize), and the Result Evaluation table plus the 'Over-researching — stop when you have enough' anti-pattern supply quality criteria and a stopping rule. Evaluation is not wired into the workflow as an inline checkpoint step, so it does not reach anchor 5's explicit validation steps. | 4 / 5 |
Progressive Disclosure | A single-file skill with no bundle files and no external references; sections are well-organized with clear headers and nothing that belongs in a separate reference file. The body exceeds 50 lines, so the simple-skill exception to anchor 5 does not strictly apply — anchor 4 ('good structure; most content appropriately placed') is the best fit. | 4 / 5 |
Total | 15 / 20 Passed |