pdfpix

Redact a PDF

or drop one here
Choose a PDF first.

Mark private content by box, by text block, or by searching for text that repeats across pages. The marked pages are rebuilt from a fresh render, so the original content stream never reaches the output. The result is then checked for the text you removed before the download appears. Nothing is uploaded.

How to redact a PDF

  1. Choose a PDF

    Press Choose a PDF, or drop one on the page. The file stays in this browser tab.

  2. Mark private content

    Draw a box over any area, select a text block, or find the same text across every page. Undo, redo and page controls stay available while you review.

  3. Choose what remains selectable

    Keep unmarked text selectable, or flatten every marked page for the strictest removal.

  4. Read the verification, then download

    The result is reopened. Marked regions, hidden text, raw PDF data and private metadata must all pass before the download appears.

Why a black rectangle is not a redaction

The common mistake is to draw a filled box over a name in a PDF editor and send the file. The box is one more drawing instruction added on top of the page. The name is still in the content stream underneath it, so anyone can select the text through the box, copy it out, or read it in a text extractor without ever seeing the page. Published documents have leaked this way many times.

This tool does not add a box. It rebuilds every page that carries a mark from scratch, and only the content you did not mark is written back into the new page. What was marked is not copied, so the marked text is permanently gone and there is nothing left underneath to recover. The same pass strips document metadata, the XMP packet, annotations, embedded files, outlines, open actions and form data from the output, because private text hides in those places too.

Then the output is read back. Every marked region is checked for black pixels, the text layer is checked for anything overlapping a mark, and the removed fragments are searched for in the raw bytes of the file. A download appears only when all of that comes back clean.

Questions

Is covering text with a black box enough?
No. This tool rebuilds every marked page, then adds back only the unmarked text when that mode is selected. The original page content is not copied into the affected page.
Are my PDF and search terms uploaded?
No. The PDF, previews, marks and search text stay in this browser tab. Anonymous analytics receive only aggregate actions and counts.
Can it find text inside a scanned image?
Not by search. Search uses the PDF text layer. Draw a box over text inside any image or scanned page to destroy those pixels.
Will unmarked text remain selectable?
Yes, in the default mode. The stricter flatten mode removes all selectable text from each marked page. Unmarked pages remain unchanged.
What hidden data is removed?
The redaction pass removes document metadata, XMP metadata, annotations, embedded files, outlines, open actions and form data from the output.
What does the redaction check cover?
It reopens every page, checks that each marked region is black, checks that no selectable text overlaps a mark, searches removed text fragments in the raw output, and scans private document data.
What is the difference between the two redaction modes?
The default mode rebuilds each marked page and puts the unmarked text back, so the rest of the page stays selectable. The flatten mode draws each marked page as an image, which removes all selectable text on that page and is the stricter choice.
Can I redact the same phrase on every page?
Yes. Search for the text once and every match across the document is marked. Review the marks before you run the redaction, because search only reaches text that sits in the PDF text layer.

The other pdfpix tools

All 30 of them run in a browser tab like this one. Each reads its own output back and checks it before offering the download, and none of them needs an account.

Where to go next