[unredact.tools]

See what your files give away.

GPS coordinates, device serials, author names, black boxes that don't actually hide anything, deleted tracked changes, printer tracking dots. Drop a file to see what it leaks, then download a copy with all of it removed.

Drop files here or click to choose

JPEG · PNG · HEIC · WebP · TIFF · PDF · DOCX · XLSX · PPTX · MP4 · MOV · MP3 · text

Your file goes straight to our server, never through the content network that speeds up these pages, and is processed in a temporary folder that is deleted before the response is sent. Nothing is stored or logged. The only thing kept is a running count of files checked and cleaned, with no details about any of them. For anything truly sensitive, run it yourself: pip install "unredact-tools[web]" && unredact.tools web

Find and remove hidden data in photos, PDFs and Office files

Photos (JPEG, PNG, HEIC, WebP, TIFF): GPS coordinates, capture time and timezone, camera make, model and serial numbers, owner names, editing software, the embedded thumbnail of the uncropped original, and motion-photo video hidden after the image. It also looks at the pixels: faces, license plates, QR codes and barcodes, readable text like addresses and card numbers, and printer tracking dots on scans.

PDFs: text still readable under black-box redactions, earlier versions of the file with the text deleted since, white-on-white, microscopic and off-page text, hidden layers, attachments, JavaScript, author and software metadata, and usernames in file paths.

Word, Excel and PowerPoint: author and company names, tracked deletions and who made them, comments, hidden text, template paths, editing session IDs, hidden sheets, rows and slides, speaker notes, and photos inside the document that keep their own metadata.

Video and audio (MP4, MOV, MP3 and more): recording location, device details and timed GPS tracks.

Text: invisible zero-width watermarks and Unicode tag characters used to tag individual copies, and look-alike letters swapped in from other alphabets.

How metadata removal works

Each clean copy is rebuilt from the visible content only, instead of deleting known fields one by one, so nothing the tool doesn't recognize gets carried over. The copy is then scanned again and the result shown before you download it. Your original file is never modified.