HTML Accessibility Checker
Find the accessibility barriers that can be detected in a page’s code. The checker runs more than twenty rules mapped to WCAG 2.2 success criteria: images and image buttons without alt text, form fields without labels, buttons and links without an accessible name, empty headings and skipped levels, a missing lang attribute or title, duplicate IDs used by labels, positive tabindex values, focusable content hidden from screen readers, invalid ARIA roles, frames without titles, data tables without headers, disabled zoom, timed refreshes and autoplaying sound.
- Encrypted connection
- No sign-up
- Free to use
How to use HTML Accessibility Checker
- Enter a page URL (or paste its HTML).
- Click “Check accessibility”.
- Fix the failed checks with the examples shown.
- Test with a keyboard and a screen reader.
HTML Accessibility Checker features
WCAG mapping
Each rule names its criterion.
Examples
The offending HTML for each problem.
Accessible names
Buttons, links and fields.
ARIA checks
Roles, aria-hidden and labelledby.
Passed checks
Listed for the record.
Paste mode
Analyse HTML you paste, e.g. a staging page.
When to use HTML Accessibility Checker
- Quick checks during development.
- Reviewing templates and components.
- Preparing for a full accessibility audit.
- Checking pasted HTML from a CMS.
HTML Accessibility Checker FAQ
Does passing mean the page is accessible?
No. Automated checks find roughly a third of the barriers. Keyboard use, focus visibility, colour contrast, meaningful text and screen-reader behaviour need manual testing.
Why is a placeholder not a label?
It disappears while typing, is often low contrast and is not reliably announced. Use a visible label element.
Does it check colour contrast?
Not in the page, because that needs the rendered styles. Use the WCAG Contrast Checker for colour pairs.
Is this a legal compliance report?
No. It is a technical aid. Legal requirements such as the ADA or the European Accessibility Act need a full review.
What the rules cover
The rules focus on failures that are both common and reliably detectable in HTML: missing text alternatives (1.1.1), missing structure (1.3.1), missing names for interactive elements (4.1.2), unhelpful link text (2.4.4), focus order (2.4.3), a page title and language (2.4.2, 3.1.1) and settings that stop people from zooming or controlling time and sound.
Each failed rule lists up to four example elements so you can find them in the source quickly. Fix the failures first; warnings such as a missing main landmark or skipped heading levels are worth fixing but do not always block access.
The checker reads the HTML as served; components that are built by JavaScript after loading are not included.
Privacy: the address you enter is fetched by our server only to run this check. The downloaded HTML is passed to your browser for analysis and is not stored, and the results are not saved on our side. Requests are rate-limited to keep the service fair for everyone, so if you check many pages in a row you may need to wait a few minutes.
Limits: pages are read up to 2 MB of HTML, redirects are followed up to eight hops and each request has a short timeout. Pages behind a login, servers that block automated requests and content added only by JavaScript cannot be analysed this way; for those, copy the page source from your browser and use the paste option where it is available.
Tip: fix the problems marked red first, because they matter most. Yellow warnings are worth reviewing but are often harmless in context, and blue notes are information. After a change, run the check again to confirm the result – search engines pick up the change on their next visit to the page.