Content
57%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 is a well-structured, appropriately split overview that correctly pushes implementation detail to references/rule.md, but the executable core is too thin in the body itself: the Check and Fix sections give direction rather than instruction, with no detection method or code example inline, and the templated Code Review section spends tokens without adding rule-specific guidance.
Suggestions
Name a concrete detection method in the Check section, e.g. the axe rule id `form-field-multiple-labels` or a DOM query for inputs whose id is referenced by more than one `label[for]`.
Inline a minimal good/bad HTML pair in the Fix section (two labels on one input vs. a single label or aria-labelledby) instead of deferring all code to references/rule.md.
Trim or merge the 'Code Review' section, which restates the description's generic boilerplate and adds no rule-specific instruction beyond Check/Fix/Explain.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean overall (no concept over-explanation), but the 'Code Review' section ('Review the rendered markup and interactive states that affect Use a single label for each form field. Flag exact elements, roles, labels, focus behavior...') is templated boilerplate duplicating the description, and 'Explain how a 1:1 relationship...' adds little rule-specific value. This fits 'mostly efficient but could be tightened' rather than 4's 'minor instances of over-explanation', since two of six sections are generic filler. | 3 / 5 |
Actionability | 'Search for form inputs that are associated with more than one <label> element' and 'Consolidate multiple labels into a single <label> or use aria-labelledby' give concrete direction, but the body provides no detection method (no selector, no axe rule id — those live only in references/rule.md) and no code example. Fits 'some concrete guidance but incomplete'; not 4 because a reviewer working from the body alone cannot execute the search. | 3 / 5 |
Workflow Clarity | The Check → Fix → Explain → Code Review sequence is clear, but validation is only hand-waved ('note how to verify the fix with browser accessibility tooling or assistive tech') with no explicit checkpoint in the body — the concrete Verification steps are deferred to the reference. Fits 'steps listed but checkpoints missing or implicit'; not 4 because the validation checkpoint is not 'mostly present'. | 3 / 5 |
Progressive Disclosure | A concise overview body with a single clearly signaled, one-level-deep reference — 'For full implementation details, code examples, and framework-specific guidance, see `references/rule.md`' — and references/rule.md exists and delivers exactly those details (code examples, best practices, verification steps). Matches the 'clear overview with well-signaled one-level-deep references' anchor; no organization gaps to justify 4. | 5 / 5 |
Total | 14 / 20 Passed |