Every WCAG 2.2 A + AA criterion — automated hard-fail detection
Not "conformance guaranteed" — we tell you exactly how each criterion is judged. All 55 Level A + AA criteria have automated hard-fail detection in the engine (80+ evaluators, pinned by three CI regression suites against a 196-page per-criterion testbed). 37 criteria hard-fail from static markup analysis alone; the rest use deterministic Playwright-driven browser probes (2.1.2 keyboard trap, 1.4.10 reflow at 320px, 2.4.7 focus visibility, measured flash rate for 2.3.1) or stability-verified vision and text judges (1.4.5 images-of-text, 1.4.1 use-of-color, 2.4.6 heading descriptiveness — criteria a rule-based scanner cannot evaluate). Cross-page aggregators check consistency across the page set (3.2.3 navigation, 3.2.6 help location), and an opt-in flow crawler proves 3.3.4 and 3.3.7 across multi-step checkout-style flows. All 86 criteria (AAA included) are evaluated with evidence; the 16 that still need human judgment are all Level AAA and flow through the in-VPAT Human Attestations table as confidence-scored verdicts for one-click reviewer signoff — so nothing is silently "passed."
