Content
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a high-level framework for accessibility auditing but lacks the concrete, actionable guidance needed to be useful. The instructions read like a project management checklist rather than executable steps — no specific tools, commands, code examples, WCAG criteria references, or output formats are provided. The content relies heavily on a referenced playbook that isn't included in the bundle, making the skill hollow in isolation.
Suggestions
Add concrete tool recommendations and executable commands (e.g., `npx axe-core`, `lighthouse --accessibility`, specific browser DevTools steps) instead of vague directives like 'Run automated scans'.
Include at least one example audit finding with WCAG criterion mapping, severity rating, and remediation code snippet to demonstrate the expected output format.
Add validation checkpoints to the workflow, such as 'Verify zero critical violations before proceeding to manual testing' and define what constitutes a passing audit.
Either include the referenced `resources/implementation-playbook.md` in the bundle or inline the most critical content (key WCAG criteria checklist, common violations with fixes) directly in the SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary framing ('You are an accessibility expert...') and context sections that restate what Claude already knows. The 'Use this skill when' / 'Do not use this skill when' / 'Context' sections add moderate bloat without much actionable value. However, it's not egregiously verbose. | 2 / 3 |
Actionability | The instructions are entirely abstract and vague — 'Run automated scans,' 'Perform manual checks,' 'Map findings to WCAG criteria' — without any concrete commands, tool names, code snippets, example outputs, or specific techniques. There is nothing executable or copy-paste ready. | 1 / 3 |
Workflow Clarity | There is a sequential list of steps (confirm scope → automated scans → manual checks → map findings → remediate → re-test), which provides a basic workflow. However, there are no explicit validation checkpoints, no feedback loops for error recovery, and the steps lack specificity about what constitutes passing or failing each stage. | 2 / 3 |
Progressive Disclosure | The skill references `resources/implementation-playbook.md` and a sub-skill, which is a reasonable structure. However, no bundle files are provided, so we cannot verify these references exist. The main SKILL.md itself is thin to the point of being unhelpful on its own, pushing essentially all substance to referenced files that may not exist. | 2 / 3 |
Total | 7 / 12 Passed |