← ClaudeAtlas

accessibility-auditlisted

axe-core through Playwright on every key screen, failing on critical and serious only, plus the seven manual checks a DOM scanner structurally cannot perform. Use when a WCAG 2.1 AA claim has to survive someone checking it — automated tooling reaches about a third of the success criteria, and the other two thirds are what actually stops a disabled user.
QAIA-Project/QAIA · ★ 0 · Testing & QA · score 70
Install: claude install-skill QAIA-Project/QAIA
# Accessibility Audit — axe-core, plus the pass axe cannot do > **Standalone adaptation.** Self-contained version of the `a11y-audit` skill from > [QAIA](https://github.com/QAIA-Project/QAIA) (MIT). The canonical version and its `references/` > live in that repository. QAIA is pre-alpha and says so. ## Naming a regulation is an implicit promise This skill's oracle is **WCAG 2.1 A/AA, and nothing else** — axe-core plus the manual pass. EN 301 549, the European Accessibility Act, France's RGAA 4.1 and Section 508 are all *built on* WCAG, which is why this run helps with every one of them. Each also adds obligations WCAG does not contain, which is why a green run is **never** a compliance verdict for any of them. **RGAA 4.1 in particular is a test *method*, not a synonym for WCAG:** 106 numbered criteria over 13 themes, a mandatory sample of pages, and a compliance rate computed as `criteria compliant / criteria applicable` over that sample. This skill reports violations by severity over the screens you point it at — **different numerator, different denominator, different sample.** No axe-core rule maps one-to-one onto an RGAA criterion, and the *déclaration d'accessibilité* is a legal artefact with a mandatory template that nothing here produces. Say this: > ✅ *"axe-core and the manual pass found no WCAG 2.1 AA violation on these screens. That covers > roughly a third of the success criteria automatically; the rest, and the RGAA method itself, > require a human audit."*