PDF Redaction — Permanently Remove Text & Data

Online tool to redact a PDF properly: black out names, passport and card numbers, addresses or any sensitive text. Type a surname or a number and every occurrence across the document is found and covered at once. Unlike a black rectangle drawn on top, a redacted page is rebuilt as a flat image, so the hidden text is physically removed from the file — it cannot be copied, searched, or extracted afterwards. Works in your browser on desktop and phone.

How to redact a PDF

Upload a PDF

Drop the file or click to browse. Each page is rendered for editing.

Find what to cover

Type a word in the search bar or press “Find data” and the boxes are placed for you. Drag over anything else by hand.

Review the boxes

The list under the page names what is covered; drop a false positive with one press, move or resize the rest.

Download the file

Covered pages are rebuilt as images with the data removed; document metadata is dropped.

Black out text and data in a document so it cannot be recovered

Drop a PDF here

PDF ·

We read the document’s own text, cover what you choose, and rebuild the file

Features

True redaction: a covered page is rasterized, so the hidden text is removed from the file — not layered over

Search the document: type a name, an account number or an address and every match on every page is covered in one press

Automatic detection reads the PDF’s own text layer: card, tax and registration numbers are validated by their checksums rather than guessed

Scans are handled too — a page with no text layer falls back to text recognition

A warning when the value you covered here is still printed uncovered on another page — the most common way a “redacted” document leaks anyway

Pages with no redaction keep their text, so the rest of a long contract stays searchable and the file stays small

Result check: the finished file is re-opened and confirmed to hold none of the covered text

What you can remove

Names, passport and ID numbers on a scanned contract
Card numbers and CVV on a statement
Home address, phone, and email on a form
Every mention of a claimant’s surname in a case file — across all pages at once
Signatures and stamps before sharing a document
Salary or price figures in an offer
A header or footer with a case number on every page

Input and output

Input: PDF (text or scanned). Output: a new PDF with no document metadata, where each redacted page is a flattened image at the resolution you choose (150, 200 or 300 dpi) and untouched pages keep their original text. Choose JPEG pages (smaller) or PNG (sharper) on export.

Why a black box in most PDF editors is not enough

In most PDF tools a black rectangle is just an annotation drawn on top — the text stays in the file and is revealed by copy-paste, search, or simply asking an AI to read under the box. Real leaks (the Manafort filing, the Mueller report, the Epstein files) happened exactly this way. Here a covered page is rebuilt as an image, so there is no text layer left to leak. The second most common failure is covering a surname on page one and missing it on page seven — this tool warns you about that itself.

Frequently Asked Questions

Can the redacted text be recovered?

No. A redacted page is rebuilt as a flat image with the covered area overwritten, so there is no text layer, OCR layer, or hidden content left underneath. This is the method security guides recommend (rasterize + flatten), unlike a black box drawn over text that leaves the text in place.

Can I cover a word everywhere in the document at once?

Yes — that is what the search bar is for. Type a surname, an account number or an address; the tool shows how many matches there are and on which pages, then covers them all in one press. It searches the PDF’s own text, so it finds occurrences you would never have scrolled to.

Why is part of the output still searchable?

Because only the pages that carry a redaction are rebuilt as images; the others keep their text. That way a 40-page contract does not lose its search over two covered paragraphs. If you want every page flattened, turn off “Leave clean pages alone” in the export settings.

What happens to the document metadata?

The output is a new PDF. The title, author, keywords, revision history, bookmarks and annotations of the original are not carried over.

What does the result check do?

After the file is built it is re-opened and read by an independent pass: rebuilt pages must hold no extractable text, and no covered value may be readable anywhere in the file. The verdict is shown right under the tool.

Does it work with scans?

Yes. A page with no text layer is passed to text recognition and the data is searched there instead. Ordinary text pages are read directly, which is faster and more accurate.

Where does the redaction happen?

The PDF is rendered, redacted, and rebuilt by JavaScript running in your browser tab, using the pdf.js and pdf-lib libraries.

Which method should I use for IDs and numbers?

The solid fill: the area is overwritten with one color and the text beneath is gone from the rebuilt page. Treat blur as a cosmetic effect only.

How many pages can I redact?

Up to 100 pages per document. For very large files, split it first. The number of exports per day is limited — the counter is shown under the tool.

How do you rate this tool?

Thank you for your rating!

Change my rating

Want to leave a comment?

Want to share more? Leave a comment!

Thank you! Your comment will appear after moderation.

Who is this tool for?

Published 11 Jun 2026 Updated 12 Jun 2026 Author: The Supreme