Redact PDF
Redact PDF removes sensitive content rather than just covering it. Draw boxes over the areas you want gone; the page is rasterised with those areas painted out and rebuilt, so the underlying text and images are genuinely destroyed and cannot be recovered by copying or searching. Because pages become images, the redacted file is no longer text-searchable.
1 Add your file
Enter the open password to use it here. It is checked in your browser and never sent anywhere.
2 Choose your options
Redaction here is real: the marked pages are rasterised with your boxes painted out and rebuilt from those images, so the original text and pictures under a box no longer exist in the file and cannot be recovered by copying, searching or extraction. The trade-off is unavoidable — a rasterised page is an image, so redacted pages are no longer text-searchable or selectable. Pages you did not mark are left as they are, with their text intact. Always open the result and check before sending it on.
How to redact PDF
Add your PDF
Each page is rendered so you can see exactly what you are removing.
Mark what must go
Drag boxes over the sensitive areas, or search for a name or number and let the tool mark every occurrence for you to check.
Download the redacted PDF
Marked pages are rebuilt with the content genuinely gone — not covered, gone.
Removal, not concealment
Drawing a black rectangle over a name adds one instruction to the end of the page: fill this area with black. The text underneath is untouched and still in the file, so it can be copied, searched or recovered by deleting the shape. This is how nearly every un-redaction story begins.
Real redaction is destructive. Text-showing operators inside the marked area are deleted from the page's instruction list and the page is rebuilt without them; where the sensitive content is inside an image, those pixels are overwritten. The black box is then drawn as a marker — the receipt, not the mechanism.
A worked example
A contract goes out with the counterparty's bank details removed. The check that proves it worked takes fifteen seconds: open the finished file, select all, copy, and paste into a plain text editor. If the account number is not in what you pasted, it is not in the document. That single test would have prevented most published un-redaction failures.
Limitations worth knowing
- Redaction does not clear document metadata. Use Edit PDF Metadata afterwards.
- Flatten first. Text inside annotations or form fields is not page content and may not be seen by a redaction pass.
- The filename travels with the document and is not covered by anything done inside it.
- Bookmark titles and link targets can leak what the page no longer says.
- Redaction protects the string, not the fact. Removing a name while leaving an unmistakable description achieves nothing.
Further reading
What PDF Redaction Really Means (And Why Black Boxes Fail), What Your PDF Quietly Reveals
Related tools
Redact PDF — frequently asked questions
Completely. A black rectangle is drawn on top of the text, which is still there underneath — select it, copy it, and it comes straight out. Newspapers and government departments have leaked documents exactly this way. Here the page is rasterised with the boxes painted out and rebuilt from the image, so there is nothing underneath to recover.
Not from the output file. The bytes are not in it. Your original file is of course untouched, so guard that separately.
Because destroying text under a box means rebuilding the page as an image, and an image has no text layer. That is the honest cost of real redaction. Pages you did not mark keep their text, and you can run OCR PDF over the result if you need searchability back.
It marks every occurrence in the text layer, which is a good start and much faster than hunting by eye. It cannot find text inside images, and it does not know that a name also appears in a scanned signature. Check the marks yourself — this is the one tool where you should not trust automation alone.
Yes. Draw boxes over the areas by hand — the pages are images already, so find-and-mark has no text layer to search. The redaction is just as permanent.
No, and this is the tool where that matters most. Rendering, marking and rebuilding all happen in this browser tab. A document you are redacting is by definition one you do not want on someone else's server.