Content
92%Weight 40%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is lean, actionable, and workflow-structured with explicit verification checkpoints and concrete cited examples. Its only gap is progressive disclosure: several referenced governance/template files are not present as actual bundle files.
Suggestions
Either provide the referenced bundle files (templates/accessibility-check.md, policies/*.md) under references/ or rephrase those mentions as conceptual pointers rather than implied local files.
Consider moving the full WCAG checklist into a references/ file and summarizing only the highest-leverage criteria inline, to better match the progressive-disclosure ideal of a one-level-deep overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and assumes Claude's intelligence, omitting basic explanations of WCAG/screen readers/ARIA; nearly every token is a cited criterion, code snippet, or anti-pattern that earns its place. | 3 / 3 |
Actionability | Provides fully executable HTML/CSS snippets, specific tool commands (axe-core via @axe-core/playwright, jest-axe, Cmd+F5), and copy-paste-ready corrections each tied to a cited WCAG criterion. | 3 / 3 |
Workflow Clarity | The "As 4 categorias de teste" table sequences the methods and "Evidencia de Conclusao" acts as an explicit validation gate (axe clean, full keyboard path, screen reader interpretation, prioritized gaps) before handoff. | 3 / 3 |
Progressive Disclosure | Sections are well organized, but referenced files (GLOBAL.md, policies/*.md, templates/accessibility-check.md) do not exist as bundle files in references/scripts/assets, so the split is signaled but not self-contained. | 2 / 3 |
Total | 11 / 12 Passed |