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, appropriately split skill body: the Check/Fix guidance is concrete and rule-specific, and progressive disclosure is exemplary with a verified, clearly signaled one-level-deep reference. The main drag is token efficiency — known-concept explanation, a redundant bullet, and generic boilerplate sections pad an otherwise lean document.
Suggestions
Drop the opening sentence explaining why the main landmark matters and the redundant Quick Reference bullet 'Provides a clear starting point for the page's primary content' — Claude already knows this; keep only rule-specific guidance.
Tighten the Explain and Code Review sections into single imperatives (e.g., 'Explain why multiple main landmarks confuse assistive-tech users') instead of full-sentence generic boilerplate.
Add a concrete counting example to the Check section, such as `document.querySelectorAll('main, [role="main"]').length`, so the check is copy-paste executable in the body itself.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The Check and Fix sections are tight, but the body includes unnecessary explanation Claude already knows — 'A single main landmark allows keyboard and screen reader users to jump directly to the primary content of the page, bypassing repetitive navigation' — plus a redundant Quick Reference bullet ('Provides a clear starting point for the page's primary content') that repeats it, and wordy boilerplate in the Explain and Code Review sections. It is mostly efficient but could be noticeably tightened, matching anchor 3 better than 4. | 3 / 5 |
Actionability | Concrete, executable instructions are present: 'Count the number of <main> elements or elements with role='main' on the page' and 'Combine multiple main content areas into a single <main> element or remove the 'role=main' attribute from extra containers.' It is not a 5 because no example selector or command for the counting check is given and the Explain/Code Review sections are generic directives rather than rule-specific guidance. | 4 / 5 |
Workflow Clarity | For this simple single-rule skill, the Check → Fix → Explain sequence is clear and the single action (verify exactly one main landmark) is unambiguous, with verification mentioned ('note how to verify the fix with browser accessibility tooling or assistive tech'). It is not a 5 because explicit validation steps are deferred to references/rule.md's Verification section rather than stated or linked in the body's own workflow, leaving minor validation gaps. | 4 / 5 |
Progressive Disclosure | The body is a concise overview and the single bundle file, references/rule.md (verified to exist), correctly holds the code examples, best practices, and verification details. The reference is well signaled — 'For full implementation details, code examples, and framework-specific guidance, see references/rule.md' — one level deep with easy navigation, matching the 5 anchor. | 5 / 5 |
Total | 16 / 20 Passed |