Content
18%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is essentially a hollow shell — it has a clear topic (screen reader testing) but provides no actionable content whatsoever. The instructions are generic boilerplate that could apply to any domain, and all substantive content is deferred to a referenced file that doesn't appear to exist in the bundle. A user or Claude following this skill would have no idea how to actually perform screen reader testing.
Suggestions
Add concrete, actionable testing steps: specific screen reader commands (e.g., NVDA: Insert+Down for read all, Tab for form navigation), keyboard shortcuts, and what to verify at each step.
Include at least one executable workflow, e.g., 'Testing a form: 1. Navigate to form with Tab, 2. Verify each label is announced, 3. Check error messages are announced on submit, 4. Verify focus management after submission.'
Add specific examples of common screen reader testing patterns: landmark navigation, heading structure verification, ARIA live region testing, and what correct vs incorrect announcements sound like.
Either provide the referenced `resources/implementation-playbook.md` bundle file or inline the essential content so the skill is usable standalone.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is relatively short but includes boilerplate sections ('Use this skill when' / 'Do not use this skill when') that add little value. The 'Do not use' section is particularly vacuous. The instructions section is generic filler rather than domain-specific guidance. | 3 / 5 |
Actionability | The instructions are entirely vague and abstract — 'Clarify goals, constraints, and required inputs' and 'Apply relevant best practices' provide zero concrete guidance on how to actually test with screen readers. No commands, no code, no specific screen reader interactions, no keyboard shortcuts, no testing procedures. | 1 / 5 |
Workflow Clarity | There is no meaningful workflow defined. The four bullet points in Instructions are generic platitudes that could apply to any skill. There are no sequenced steps, no validation checkpoints, and no specific testing procedures for screen reader testing. | 1 / 5 |
Progressive Disclosure | The skill references `resources/implementation-playbook.md` for detailed patterns, which is a reasonable structure, but no bundle files were provided to verify the reference exists. The SKILL.md itself contains almost no substantive content — it delegates everything to a file that may not exist, making the skill useless on its own. | 2 / 5 |
Total | 7 / 20 Passed |