An accessible EPUB tells reading systems what it offers (through accessibility metadata), declares its language, has a working table of contents, uses real headings, gives images a text alternative and marks up tables and links so a screen reader can use them. This checker opens an EPUB in your browser, reads the package, the navigation document and every content document in reading order, and lists the technical findings it can detect, with the WCAG success criterion each one relates to and the file it is in.
It is an accessibility preflight. Its findings are technical and separate from legal compliance: it does not claim that a book conforms to WCAG or satisfies the European Accessibility Act. The EPUB is not uploaded.
Loading the tool…
Overview
About EPUB Accessibility Checker
Metadata. EPUB Accessibility 1.1 asks for discovery metadata in the package document: schema:accessMode (for example textual, visual), schema:accessModeSufficient, schema:accessibilityFeature (alternativeText, tableOfContents, structuralNavigation and so on, or "none"), schema:accessibilityHazard (none, flashing, motion simulation, sound, or unknown) and schema:accessibilitySummary. The checker reports what is missing, shows any conformance statement (dcterms:conformsTo) and its certifier, and compares the declared features with the content: images with alt text but no "alternativeText" declared, "alternativeText" declared while images have no alt attribute, MathML without the MathML feature, a page list without pageNavigation, a table of contents without tableOfContents. A conformance claim next to errors is flagged as unsupported by these results.
Package and navigation. Language (dc:language, valid BCP 47), title, identifier, modified date, spine, page progression direction for right-to-left languages, the navigation document with a table of contents (empty links, links to files that are missing), page list when a print source is named, landmarks, EPUB 3 versus EPUB 2 (the specification targets EPUB 3), fixed-layout publications, and the container rule that the mimetype file comes first and stored.
Content documents. Per document: language, title, heading structure (skipped levels, empty headings, none in long texts), images without alt attribute, alt text that is a file name, decorative images marked alt="" , SVG without an accessible name, tables without header cells, empty and generic links such as "click here", MathML without alternatives, videos without captions, audio, moving text, positive tabindex, unlabeled form controls, page-break markers without labels, scripted content (and whether the manifest declares it), pages made of pictures only, and duplicate ids. Documents with a named entity or an external DTD are handled safely.
Separate from legal compliance. The report ends with what a program cannot judge: reading order, contrast, whether alt text is useful, language changes, and most of WCAG. The European Accessibility Act has covered e-books since 28 June 2025; what that means for a publisher is a legal question this tool does not answer.
Compatibility
Supported Formats
EPUB files (EPUB 3 mainly; EPUB 2 gets the checks that apply to any XHTML) up to 150 MB with up to 400 content documents. Not supported: Kindle formats (AZW, KFX, MOBI), encrypted (DRM) EPUBs, audio-only publications and EPUB files that are not valid ZIP packages.
Guide
How to Use EPUB Accessibility Checker
- Drop an .epub file onto the box (up to 150 MB).
- Read the summary, then the accessibility metadata that the package declares.
- Go through the findings by area: metadata, navigation, structure and content. Each shows the file and the WCAG criterion.
- Fix the source files (or your publishing tool), rebuild the EPUB and check again. Download the findings as CSV or JSON.
Highlights
Features
- Accessibility metadata The five discovery properties, the conformance claim and a comparison with the content.
- Navigation and structure Table of contents, page list, landmarks, language, spine and package rules.
- Content checks Headings, alt text, tables, links, MathML, media, scripts, forms and page-break markers.
- WCAG references and files Every finding names the document and the related success criterion.
- Technical, not legal Clearly separated from conformance and legal compliance; CSV and JSON export.
Applications
Use Cases
- Checking an EPUB for missing accessibility metadata before it goes to a retailer
- Finding images without alt text across a whole book
- Seeing whether the navigation document and headings give a usable structure
- Preparing for an accessibility review by listing the technical issues first
- Comparing what the metadata claims with what the content shows
Good to Know
Limitations
This is an accessibility preflight of technical findings. It does not claim full WCAG conformance, does not assess legal compliance (including the European Accessibility Act), and a report with no errors is not a conformance statement. A program cannot judge reading order, colour contrast, the quality of alt text, language changes, whether content makes sense when read aloud or most of WCAG; that needs a review by a person with a screen reader and accessibility expertise.
The metadata checks follow EPUB Accessibility 1.1 (a W3C Recommendation; a newer revision may exist, so check the W3C site). Content checks run on the XHTML as written; scripts are not run, so dynamic content is not assessed. Documents above 6 MB and beyond the first 400 are not inspected.
This tool runs entirely in your browser — your file is never uploaded to our servers. See our Privacy Policy for full details.
Support
Frequently Asked Questions
Is my EPUB uploaded?
No. The file is opened and checked in your browser and is not sent to our server.
Does a clean report mean my EPUB is WCAG or EAA compliant?
No. The checker finds technical issues a program can detect. It cannot judge reading order, contrast or the quality of alt text, and it does not assess legal compliance. A person must review the publication.
Which accessibility metadata is required?
EPUB Accessibility 1.1 asks for schema:accessMode, accessModeSufficient, accessibilityFeature, accessibilityHazard and accessibilitySummary, and recommends a conformance statement with dcterms:conformsTo. The checker reports each one that is missing.
What is the difference between alt="" and a missing alt attribute?
alt="" marks a decorative image that a screen reader skips. A missing alt attribute is an error: the reader may announce the file name.
Does it check EPUB 2 files?
Partly. EPUB Accessibility 1.1 is written for EPUB 3, so for EPUB 2 the content checks run but the navigation and metadata expectations are noted as not applicable.
What does the European Accessibility Act have to do with this?
It has covered e-books since 28 June 2025. This tool reports technical findings that relate to accessibility, and it makes no statement about legal compliance, which depends on national law and expert assessment.
Keep Going
Related Tools
- PDF/A Checker & Converter Check which PDF/A level a PDF declares and spot common problems in your browser. Convert to PDF/A-1b, -2b or -3b on the server if you need to.
- PDF Accessibility Checker A quick technical preflight of a PDF: tagged or not, language, title, headings, alt text, form labels, bookmarks. Not a certification. In your browser.
- Screenshot Color Contrast Checker Pick the text and background colours from a screenshot (or type them) and see the WCAG contrast ratio with pass or fail for each level. In your browser.
- Office Document Inspector & Metadata Remover Remove metadata from Word, Excel and PowerPoint files: see author names, comments, tracked-change authors, hidden sheets and speaker notes first, then clean a copy. In your browser.
- DOCX Heading & Numbering Structure Checker Check the heading structure of a Word document: skipped levels, fake bold headings, empty headings, inconsistent or duplicate numbering and an out-of-date table of contents. In your browser.