Content
75%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-organized, actionable reference that largely assumes Claude's competence and uses tables effectively. Gaps are minor: a couple of trimmable explanations, a script path with no matching bundle file, and the absence of an explicit numbered workflow.
Suggestions
Remove or trim the 'Large text per WCAG' definition and the tooltip-hoverable failure narrative, since Claude already knows WCAG terminology.
Either create scripts/check-contrast.js in a bundle or reframe the code as an inline snippet without the implied external file path.
Convert the design-token bulk-check into a short numbered workflow (define pairs -> run script -> treat non-zero exit as a failing CI gate) to make the validation loop explicit.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly lean reference material (ratio tables, failures, anti-patterns) with a few minor over-explanations such as the 'Large text per WCAG' definition and the tooltip-hoverable failure narrative that could be trimmed. | 4 / 5 |
Actionability | Provides executable JS (check-contrast.js), tool URLs, and concrete fix patterns ('Darken to #767676'); the script's require of ../tokens/colors.json is a relative path that may not resolve, a minor gap. | 4 / 5 |
Workflow Clarity | The bulk token check is a batch operation with a validation loop (violations -> console.error -> process.exit(1)) and CI wiring via a11y-violation-gate; sequence is clear though not a numbered multi-step checklist. | 4 / 5 |
Progressive Disclosure | Well-sectioned single SKILL.md with one-level-deep references to peer skills; the inlined 'scripts/check-contrast.js' path implies a bundle file that is not present (no scripts/ dir), a minor organization gap. | 4 / 5 |
Total | 16 / 20 Passed |