PDF Accessibility Checker
A PDF accessibility checker tests whether a PDF carries the structure assistive technology needs: tags, a heading hierarchy, alt text on figures, header cells in tables, a document language and title, bookmarks for navigation, and a real text layer rather than a scan. This one runs in your browser against the machine-checkable requirements shared by WCAG 2.1 Level AA and PDF/UA-1, cites the specific clause for every finding, and works on any operating system — the established free checker, PAC, is Windows-only desktop software.
Runs locally — your file is never uploadedNo account, no install, no page limit
Choose a PDF to check
What this checks
Ten structural requirements that can be decided from the file itself, each mapped to the success criterion or PDF/UA clause it comes from:
- A real text layer, not a scan
- Tagged structure on every page
- Heading hierarchy with no skipped ranks
- Alt text on every figure
- Header cells in every table
- Document language set
- Document title set and shown
- Bookmarks for navigation
- Encryption that permits extraction
- Pages that claim tags but have none
What it cannot check
Reading order can be wrong while every tag is present. Alt text can exist and describe nothing. Colour contrast is a rendering property, not a structural one. No automated tool decides these — not this one, not Acrobat's Full Check, not PAC — and roughly half of WCAG cannot be machine-tested at all. Treat a clean report as "nothing structural is obviously broken", then tab through the document and read it with a screen reader.
Frequently asked questions
Is my PDF uploaded anywhere?
No. The check runs entirely in your browser — the file is read locally and never sent to a server. That matters if you work for a government body, school or law firm whose documents cannot legally be handed to a third-party service just to be tested.
Does passing every check mean my PDF is WCAG compliant?
No, and no tool can tell you that. This checks the machine-testable subset: tagging, headings, alt text presence, table headers, language, title, bookmarks and the text layer. Reading order, whether alt text is meaningful, and colour contrast all need a human. Roughly half of WCAG cannot be automated by anyone.
What standard does this check against?
The structural requirements shared by WCAG 2.1 Level AA and PDF/UA-1 (ISO 14289-1), which is the pairing the ADA Title II rule and Section 508 both point at for PDFs. Every finding cites its specific success criterion or clause so you can put it in a remediation ticket.
When do US public bodies have to comply?
The DOJ ADA Title II web rule covers electronic documents including PDFs. In April 2026 the deadlines were extended by a year: entities serving populations of 50,000 or more must conform to WCAG 2.1 AA by 26 April 2027, and smaller entities by 26 April 2028.
Why does it say my document is not tagged when Acrobat says it is?
The most common cause is a /Marked flag with no structure tree behind it — a tagging step that failed part-way, or a later tool that stripped the tree and left the flag. This checker reports what it can actually read on each page rather than trusting the flag, so it will tell you the two disagree.
Can it fix the problems it finds?
Partly. Missing bookmarks it can fix — the TOC generator builds an outline from your document’s headings. Tagging, alt text and reading order have to be fixed in the source document or a full remediation tool; this page tells you precisely what is wrong and cites the clause, but it does not pretend to auto-remediate.