PDF Accessibility Inspector
Review PDF titles, language, page structure, figures, form labels and extracted reading order. Preview pages, record manual findings and export an accessibility review report.
Choose a document to begin. This tool does not certify compliance.
Extracted content order
Diagnostic extraction; compare manually with visual and tagged order.
No page loaded.
Exposed structure roles
No structure loaded.
Links and form fields
Using PDF Accessibility Inspector
Review PDF titles, language, page structure, figures, form labels and extracted reading order. Preview pages, record manual findings and export an accessibility review report.
- Open a PDF or try the example. Run the inspection and review document findings.
- Select a page to compare its artwork with extracted text and structure. Review headings, figures, links and form labels.
- Complete the manual checklist, add notes and download your JSON or text report. Fix issues in the source document and inspect a new export.
Example
Try the two-page review example. Its missing structure tags demonstrate why selectable text alone does not establish accessibility.
Questions & answers
Does this certify PDF/UA or WCAG compliance?
No. This is a screening and manual review assistant, not a compliance validator. Tags, text and labels can exist while being incorrect. Screen-reader testing and a standards-aware review are still needed.
Can it repair the PDF?
It exports findings and notes, not a repaired PDF. Correct reading order, alternative text and tagging in an authoring or PDF remediation application.
Why does extracted reading order differ from the page?
Text extraction follows content-stream order, which can differ from visual or semantic order. The displayed text is a diagnostic aid, not a simulation of every screen reader.
Can scanned pages be checked?
Pages with no extracted text are flagged for manual review and may need OCR. OCR alone does not supply correct document structure.
What is inspected?
Document title and language, declared tagging, page structure roles, figures without exposed alternate descriptions, unlabeled form widgets and links without exposed labels. Some accessibility data is not exposed by PDF.js; findings require confirmation.
What limits apply?
One PDF up to 32 MiB and 100 pages. Inspection can be cancelled; an interrupted report is clearly marked partial.
Are documents uploaded?
Parsing and preview run in your browser. Advertising and optional analytics are described in the site privacy policy.
Related tools
A useful next step for your task.
PDF ToolsPDF to Text ExtractorRecommendedExtract selectable text from PDF pages locally. Choose page ranges, review line breaks, unlock password-protected files and download TXT or page…
Developer & DataPDF Metadata EditorInspect and edit title, author, keywords and dates in a static PDF copy. Clear standard document properties locally and download a metadata report.
PDF ToolsPDF Editor & Form FillerRecommendedEdit a PDF locally: fill forms, add text, images, drawn signatures, highlights and shapes. Create new fillable fields, undo edits, save plans and…
PDF ToolsPDF Merge & SplitHighly recommendedMerge PDFs in order, extract selected pages or split them into individual PDFs locally. Page ranges, file limits, downloads, instructions and Q&A.…
Help improve this tool
Report a problem or suggest an improvement
Describe the issue without pasting private tool input. Feedback goes to our admin inbox.