Document Accessibility
PDF, Word, PowerPoint, and video - uploaded and tested
Upload a file directly from your dashboard and get real, structural accessibility checks back - no crawl, no cloud-storage sync required. Every check below genuinely parses the file; nothing is estimated or simulated.
Checks whether the PDF is tagged at all, has a document title and declared language, whether form fields have an accessible name (a tooltip, not just an internal field ID), and whether pages have an extractable text layer - the most common sign of an un-OCR'd scan. These are structural presence checks: a tagged PDF can still have the wrong reading order or a missing alt-text description on a specific image, which this doesn't verify.
Word and PowerPoint
Word (.docx) files are checked for a title, a declared language, images missing alt text, whether any heading styles are used at all, and vague link text like "click here". PowerPoint (.pptx) files are checked for a title, images missing alt text, vague link text, and slides missing a real title placeholder (not just text that looks like a title). Excel and the older binary .doc/.ppt formats aren't supported yet.
Video (MP4/MOV)
The narrowest of the four - there's no PDF/UA- or Office-equivalent structured accessibility metadata for video, so this only checks whether the file's container declares a caption/subtitle track at all, and flags a video with no audio track for manual review. It does not check caption accuracy or sync, does not detect audio description, and doesn't yet support WebM or other non-MP4 containers.
What this doesn't do yet
Documents and video are scanned by direct upload only right now - there's no automatic discovery during a website crawl, and no SharePoint, OneDrive, or Google Drive sync. Scanned/image-only PDFs are flagged, not OCR'd - the original file is never modified. All of this is on the roadmap, not silently assumed to exist.