Content
60%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 lays out a serviceable research workflow with some tool-specific details (allowed_domains/blocked_domains) and explicit verification steps, but about a third of it is generic filler and the guidance lacks concrete artifacts such as example queries. Verification ('Verify key facts across multiple sources') and contradiction tracking keep workflow clarity solid despite the vague spots.
Suggestions
Delete the boilerplate 'Use this skill when', 'Do not use this skill when', and generic 'Instructions' sections — they restate the frontmatter and add no actionable guidance, tightening conciseness.
Add a concrete example query showing the techniques in action (e.g., '"exact phrase" -unwanted site:example.com') and show the actual allowed_domains/blocked_domains syntax so the strategies are executable rather than descriptive.
Add an error-recovery step to the Approach, e.g., 'If initial queries return weak results, reformulate with broader terms or different operators before concluding the information is unavailable.'
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The guidance sections (Search Strategies, Approach, Output) are lean lists, but roughly a third of the body is generic boilerplate — 'Working on search specialist tasks or workflows', 'Apply relevant best practices and validate outcomes' — that teaches Claude nothing new. Fits 'mostly efficient but includes some unnecessary explanation or could be tightened'; not 2 because it never explains concepts Claude already knows. | 3 / 5 |
Actionability | Some concrete guidance exists — 'allowed_domains for trusted sources', 'blocked_domains to exclude unreliable sites', 'Create 3-5 query variations', 'Use specific phrases in quotes for exact matches' — but there are no example queries or executable commands, and steps like 'Target specific timeframes for recent/historical data' and 'Parse structured data from pages' omit key details. Fits 'some concrete guidance but incomplete; missing key details'; not 4 because the gaps exceed minor. | 3 / 5 |
Workflow Clarity | The Approach section gives a clear 5-step sequence ('Search broadly first, then refine' → 'Verify key facts across multiple sources' → 'Track contradictions and consensus') with an explicit verification step, fitting 'clear sequence with most checkpoints present'. Not 5 because there is no error-recovery guidance for weak results or conflicting sources. | 4 / 5 |
Progressive Disclosure | Well-organized sections with headers and appropriately inline, self-contained content; no bundle files exist and none are referenced, so nothing is misplaced. Minor organization gaps from the boilerplate 'Use this skill when' / 'Do not use this skill when' sections keep it at 'good structure; minor organization gaps' rather than a fully clean 5. | 4 / 5 |
Total | 14 / 20 Passed |