Adobe Acrobat Pro's accessibility checker costs $239/year and only catches the obvious failures. PAC 3 is free but checks PDF/UA-1 structure only — it won't catch contrast failures, missing font maps, or broken form labels. Axe PDF sends your documents to their servers. PDF Auditor runs 8 compliance check categories entirely on your machine — completely free, no uploads, no subscription.
WCAG 1.3.1 / PDF/UA-1 require a tagged PDF with a /MarkInfo dictionary so screen readers can parse heading hierarchy, lists, tables, and reading sequence. PDF Auditor flags untagged PDFs as a Critical finding — the single most common barrier to AT access.
A PDF can be visually correct but deliver content to screen readers in the wrong sequence — common in multi-column layouts, scanned documents, and PDFs exported from design tools. PDF Auditor checks marked-content sequences on every page in parallel to detect reading order breaks.
Long documents without a /Outlines bookmark tree force screen reader users to navigate linearly — violating WCAG 2.4.5. PDF Auditor flags missing outlines on multi-page documents and reports them as a Minor finding with the relevant criterion.
Screen readers need a /Lang entry in the PDF catalog to select the correct pronunciation engine. Missing language is a Critical finding under WCAG 3.1.1. PDF Auditor checks the catalog root — not just the XMP metadata, where tools often place it incorrectly.
PAC 3 doesn't check contrast. Acrobat Pro's checker only flags obvious cases. PDF Auditor extracts foreground and background color pairs directly from the page content stream and computes relative luminance per the WCAG 2.x spec — 4.5:1 for normal text, 3:1 for large text — on every page in parallel.
Every Figure element in a tagged PDF must carry an /Alt attribute or an associated ActualText entry. PDF Auditor checks both paths — catching images where the tag exists but the text is empty, a pattern Acrobat's checker misses. Results distinguish decorative images from informational ones.
Fonts without ToUnicode maps produce unreadable characters when copied to clipboard or read by a screen reader — the text renders visually but AT receives garbage. PDF Auditor checks every font resource for both embedding completeness and ToUnicode presence, reporting each violation with the font name.
AcroForm fields without /TU (tooltip) or associated label tags are invisible to screen readers — users cannot know what a field asks for. PDF Auditor walks the AcroForm dictionary and flags every unlabeled interactive field with the field name for quick remediation.
Link annotations that say only "click here" or carry no description fail WCAG 2.4.4. PDF Auditor checks every URI and GoTo annotation for accessible descriptions — not just whether an annotation object exists, but whether its /Contents or /TU field provides meaningful context.
A PDF/UA-1 conformance claim requires a pdfuaid:part=1 entry in the XMP metadata stream. PDF Auditor inspects the catalog's /Metadata stream directly — not just the document info dictionary — so it catches the cases where Word/InDesign exports place metadata in the wrong location.
PDF Auditor collapses findings across all 8 categories into a single 0–100 accessibility readiness score. Issues are deducted by severity — Critical violations cost 15 points each, Major 8 points, Minor 3. The score maps to a letter grade so you can immediately communicate compliance status to stakeholders without reading a 40-line report.
Every byte of your PDF is processed on your hardware. No file upload, no API call, no cloud dependency. For legal documents, healthcare forms, financial reports, or confidential contracts — that's not a feature, it's the only acceptable architecture.
Per-page checks (contrast, reading order, images, links) run in parallel across all CPU cores via Rayon. A 300-page report PDF audits in seconds, not minutes. Repeated issues across pages are collapsed into a single finding so you get a focused report, not 300 identical entries.
Every audit writes a structured JSON report for programmatic processing or CI/CD integration — and a human-readable HTML report that opens in the browser automatically. Both include the score, grade, per-finding severity, affected pages, and the specific WCAG or PDF/UA-1 criterion violated.
Use the native Windows GUI — drag a PDF onto the drop zone and get an instant visual summary. Or integrate the CLI into a document processing pipeline with pdf_auditor file.pdf --html. The same binary, both modes, no extra tools required.
Download, drag a PDF, read the report. No login, no API key, no setup wizard. If your PDF has accessibility issues, you'll know exactly what they are, where they are, and which WCAG criterion each one violates.