Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill is well-structured with a clean progressive-disclosure split to a real reference file, but the body's guidance is short on concrete mechanisms and explicit verification steps.
Suggestions
Add one concrete fix mechanism in the body — e.g. 'add scope="col"/"row" or link via headers="id"' — so the Fix step is actionable without opening the reference.
Make verification explicit in the workflow, e.g. 'Confirm each <th> maps to ≥1 <td> in the accessibility tree before closing the review.'
Trim redundancy: fold the Quick Reference bullets into Check/Fix or drop the meta 'Explain' section to tighten token use.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and avoids explaining basics, but the Quick Reference bullets overlap with the Check/Fix sections and the 'Explain' section is meta guidance that could be trimmed. | 2 / 3 |
Actionability | It states concrete rules ('Every <th> must be associated with one or more <td> cells') but the how-to is vague — 'ensure they are correctly mapped to data cells' gives no mechanism (scope, headers/id) and the body carries no example. | 2 / 3 |
Workflow Clarity | A Check→Fix→Code Review sequence exists, but verification checkpoints are only implicit in the body (the actual validation steps live in the reference), so it is not a fully validated workflow. | 2 / 3 |
Progressive Disclosure | The body is a clear overview with a single well-signaled one-level-deep pointer to references/rule.md (which exists), matching the clean-split anchor. | 3 / 3 |
Total | 9 / 12 Passed |