Website Accessibility Checker — WCAG 2.1 AA, tested in a real browser

Why a real browser makes a better accessibility checker

Many free accessibility checkers read your HTML and never render it. That catches a missing alt attribute, but not the problem WebAIM finds on more home pages than any other year after year: text that does not have enough contrast against its background. Contrast depends on the computed colour of the text and of whatever sits behind it once every stylesheet has applied, and an HTML-only tool never sees that. The same goes for content that only exists after JavaScript runs — menus, product grids, forms rendered by a framework.

This checker loads the page in a real browser — Monity's monitoring engine — waits for its stylesheets and scripts, and then runs axe-core, the open-source rules engine from Deque that also powers the accessibility audits in Google Lighthouse and Microsoft Accessibility Insights, against the page exactly as a visitor would get it. Our SEO audit runs axe-core too, but on the raw HTML crawlers read, so it deliberately skips contrast; this tool is the one that tests it.

What it tests

Every axe-core rule tagged for WCAG 2.0 and 2.1 at levels A and AA — the level the laws reference. Best-practice rules that are not WCAG requirements are left out, so every failure maps to a success criterion. In practice that covers:

• Text alternatives — images, image buttons and other non-text content without an accessible name (WCAG 1.1.1).
• Colour contrast — text below 4.5:1 against its background, or 3:1 for large text (1.4.3).
• Names, roles and values — buttons, links, form fields and frames that assistive technology cannot name, and ARIA attributes or roles that are invalid or misused (4.1.2).
• Forms — inputs without labels, and autocomplete values browsers cannot use to fill them in (4.1.2, 1.3.5).
• Structure — lists and tables marked up so their relationships survive a screen reader (1.3.1).
• The document — a page title and a valid language, which screen readers use to pick a voice (2.4.2, 3.1.1).

How to read the results

Each finding is one rule, with the number of elements that fail it, CSS selectors to find them, and Deque's guidance on the fix. Rules that axe rates critical or serious are listed as "to fix"; moderate and minor ones as "worth knowing". The accessibility score starts at 100 and loses points per failing rule, weighted by that severity — useful for comparing a page with itself after a fix, not as a grade of conformance. Failures in shared templates (headers, footers, product cards) are the best value to fix first, because one change repairs every page that uses them.

What automated testing can't tell you

Rules can confirm that an image has alt text, not that the alt text describes it; that a button has a name, not that the name makes sense. Estimates of how much of WCAG automated testing covers vary with how you count, but around a third of issues is the figure most often quoted. The rest takes a person:

• Use the page with only a keyboard: everything reachable, in a sensible order, with a visible focus indicator.
• Zoom to 400% or narrow the window to 320px: content should reflow without horizontal scrolling.
• Submit a form with mistakes: errors should be described in text and tied to the field.
• Listen with a screen reader — VoiceOver and NVDA are free — for unlabeled controls and dialogs that trap focus.
• Check captions on video and a way to pause anything that moves.

Fixing colours? The color contrast checker tests any pair against WCAG AA and AAA and suggests the nearest colour that passes.

WCAG, the ADA, Section 508 and the European Accessibility Act

The laws differ, but they point to the same technical standard. The European Accessibility Act, which has applied since 28 June 2025, relies on EN 301 549, whose web requirements are WCAG 2.1 level AA. In the United States, the Department of Justice's 2024 rule for state and local government websites under ADA Title II adopts WCAG 2.1 AA, Section 508 references WCAG 2.0 AA for federal agencies, and courts in ADA Title III cases about private businesses routinely look to WCAG as well. WCAG 2.2, published in 2023, adds nine success criteria, most of which need manual testing; a Monity monitor can run the WCAG 2.2 AA rules axe-core has for them.

No automated tool can certify compliance with any of these — not this one, and not an overlay widget, which changes nothing in the code assistive technology reads. A scan like this is the first pass of an audit, and the cheapest way to make sure fixed problems stay fixed.

Keep it accessible after every deploy

Accessibility rarely breaks in a redesign everyone reviews. It breaks in an ordinary release: a new component without labels, a colour tweak in the design system, a third-party widget. Monity runs this same check on the pages it monitors and alerts you in Slack, Teams or email when new WCAG failures appear or an existing one spreads to more elements — so the regression is caught the day it ships, with the evidence of when it started.

For the rest of the page, the website speed test measures how fast it loads, and the SEO audit covers headings, metadata and what search engines and AI assistants can read.