Content
71%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.
A well-structured, lean instruction skill with a concrete workflow, explicit source lists, and a fully specified output format. Its weaknesses are duplicated rules across the Contract and Rules sections, the absence of an explicit validation step in the workflow, and body references pointing to bundle files that do not exist.
Suggestions
Fix or remove the dangling references: ../index/SKILL.md, ../../references/critical-overrides.md, and ../user-context/SKILL.md do not exist in the skill bundle, so either ship those files or drop the pointers.
Deduplicate the Contract and Rules sections: "Do not overclaim from anecdotes" appears verbatim in both, and "Cite sources wherever available" / "Use citations wherever available" are the same rule stated twice; merge them into one section.
Add an explicit validation checkpoint to the workflow (e.g., after clustering, verify each problem has at least two independent sources and mark single-source problems as low confidence) to close the feedback-loop gap.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and imperative with no concept explanations Claude already knows, but rules are duplicated between the Contract and Rules sections ("Cite sources wherever available" vs "Use citations wherever available", and "Do not overclaim from anecdotes" appears verbatim in both). These minor redundancies could be trimmed, matching the efficient-with-minor-excess anchor rather than the every-token-earns-its-place anchor of 5. | 4 / 5 |
Actionability | As an instruction-only skill it gives concrete guidance: named public sources (Reddit, X/Twitter, Hacker News, Stack Overflow, GitHub issues), named internal sources (Slack, Gong, Notion, Drive, Linear/Jira), explicit clustering categories, ranking criteria (severity, frequency, confidence, product leverage), and a fully specified output template with an executive read of 5-7 sentences and three opportunity buckets. Minor gaps remain, such as how to formulate searches or how to gauge frequency signal, so it stops short of fully executable guidance. | 4 / 5 |
Workflow Clarity | The seven numbered steps form a clear, coherent sequence from scope restatement through source searching, clustering, ranking, and storytelling, with partial checkpoints via the Contract ("Say clearly when source access is missing or weak", "Separate observed evidence from inference"). However, there is no explicit validation or feedback step in the workflow itself, so it matches "clear sequence with most checkpoints present; minor validation gaps" rather than the explicit feedback-loop anchor of 5. | 4 / 5 |
Progressive Disclosure | Sections are well organized and the inline content is appropriately sized for a research skill, but all three file references in the body ([../index/SKILL.md], [../../references/critical-overrides.md], [../user-context/SKILL.md]) point to files that do not exist in this bundle, and no references/ or scripts/ files are provided. Navigation is therefore broken for the referenced materials, fitting "references present but not clearly signaled / could be better organized" rather than the good-structure anchor of 4. | 3 / 5 |
Total | 15 / 20 Passed |