Sanitise PDF (Remove Hidden Data)

A PDF carries much more than the pages you see. The author’s name and the software they used, earlier versions of the text kept by “quick save”, scripts that run when the file opens, files tucked inside it, comments from reviewers, layers that are switched off, small preview pictures of every page, and private editing data that apps like Acrobat and Illustrator leave behind. Sanitise PDF finds all of it and lets you remove it before you send the file to a client, a court, a journalist or the whole internet.

It starts by scanning each file and showing you, with real counts, what it found: the document properties with their values, every script and where it runs, the names and sizes of attached files, links that open other files or programs, hidden layers by name, and more. It also checks the pages for text a reader can’t see, such as white text or words placed off the edge of the page, and reports it so you can deal with it.

You tick what to remove; the defaults take out everything that doesn’t change how the pages look. The tool then writes a fresh copy, drops every object nothing points to any more, and scans the new file again, so the before → after table shows what really went. Your files are processed in your browser and never uploaded.

1. Tap “Choose PDFs” and pick one file, or up to 30.
2. Read the list under “What’s hidden”, beside a large view of the pages; pages with hidden text are labelled on their thumbnails. Tap a file’s tab, or its name in “All files”, to see its own findings.
3. Tick what to remove. “Safe choices” keeps everything you can see on the page; “Tick all” removes the lot. For form fields, choose “Clear the answers” or “Remove the fields”.
4. Press “Clean PDF”. Each file is cleaned and scanned again.
5. Check the before → after table, then press “Download clean PDF” (a ZIP when there are several files).

• Finds document properties and XMP metadata, JavaScript (document scripts, open actions, link, button, page and field triggers), attached files, launch and other-file links, form-submit actions, web links, comments, form fields and answers, hidden layers, page thumbnails, private app data and bookmarks.
• Shows real counts and details: property values, script locations, attachment names and sizes, layer names.
• Reports hidden text: white text, invisible text, text off the page and text too small to read.
• Always removes earlier saved versions and leftover objects, so deleted content can’t be dug back out.
• Hidden layers are taken out of the page itself, not just switched off.
• Before → after table for every file, from a fresh scan of the cleaned copy.
• Batch: up to 30 PDFs cleaned the same way, downloaded as one ZIP.
• The clean file gets no properties of its own: not even a “made with” note.

• Sanitising doesn’t touch the visible words on the page. To remove names or numbers people can read, use Redact PDF first, then sanitise.
• Hidden text is reported, not removed: white text on a dark box is often normal. If it’s a problem, redact it or remove it in the app that made the file.
• A digitally signed PDF loses its signature when anything in it changes. Sanitise before you sign, not after.
• Password-protected PDFs can’t be read. Remove the password with Unlock PDF, sanitise, then protect the clean copy again.
• If a layer’s hidden content can’t be taken out safely, the tool keeps that layer hidden and tells you, rather than revealing it.
• Clearing form answers leaves an empty form someone can fill in again; removing the fields takes the boxes off the page too.