Redact PDF
Permanently remove sensitive text and graphics — the content is destroyed, not just covered.
🔒 Privacy: processed locally in your browser — your files never leave your device.
or drop your PDF here
Find sensitive content
Search the document's text, or run a detector. Every match is listed with a checkbox — only the ticked matches become marks when you press the accept button.
Marked for redaction
Nothing marked yet. Draw areas with the Redact tool, mark a whole page, or accept matches from the detection panel.
Output quality
Redaction rebuilds every page as an image at this resolution. Higher DPI keeps small text crisper but makes a bigger file. The rebuilt document has no text layer at any setting — that is what makes the redaction permanent.
Your redacted PDF is ready!
Why redact with PdfBleu
Destroyed, not covered
Tools that draw a black box over live text leave the words inside the file — anyone can copy them out. Here every page is rebuilt as an image, so what is under the black simply no longer exists.
Find it automatically
Search for any text, or let the detectors sweep the document for card numbers (checksum-verified, so look-alike numbers are ignored), phone numbers and email addresses.
Verified before download
Before you download, the finished file is re-opened and searched on your device: if a marked item could still be found in it, you would see a warning instead of a download.
How to redact a PDF
Open your PDF
Click the select button or drop your PDF into the area provided. All of its pages open in the editor, right here in your browser.
Mark what must go
Drag black areas over sensitive content with the Redact tool, mark a whole page, or use search and the detectors to find card numbers, phones and emails in seconds.
Review the marks
Every mark is listed in the “Marked for redaction” panel, grouped by page. Remove any mark with its ✕, or clear them all and start the marking again.
Redact and download
Press “Redact document”. The pages are rebuilt as images with the marks burned in, the result is verified, and your redacted PDF is ready to download.
Frequently asked questions
Why rebuild the pages as images? Why not just draw black boxes?
Because a black box drawn over live text is not redaction — the words are still inside the file, unchanged. Anyone can select the text under the box, copy it, or remove the box and read everything. That failure has embarrassed governments and law firms in real published documents. Rebuilding each page as an image with the marked areas painted black before the image is created means the covered content is never written into the new file at all: there is nothing left to recover.
What happens to the pages and parts I did not mark?
They are rebuilt the same way. Every page of the output is a fresh image of the original page, so no hidden text, metadata, annotations, form fields or layers survive anywhere in the document — not just under the black areas. The new file is a clean, flat document that shows exactly what the original showed, minus what you marked.
Can I redact a scanned PDF?
Yes — drawing areas by hand works on any page, scanned or not, because the rebuild works on the page picture itself. What does not work on a scan is automatic detection: a scanned page has no text layer for the search or the detectors to read. The panel tells you when that is the case, and running the file through OCR PDF first adds a text layer that makes detection work.
What exactly do the detectors find?
The text search finds every occurrence of the words you type, on every page. The card detector finds digit sequences of 13–19 digits that pass the Luhn checksum — the mathematical check card numbers are built to satisfy — so random long numbers that merely look like cards are not flagged. The phone detector finds 7–15 digit numbers with the usual separators and optional country codes, and the email detector finds standard email addresses.
Does the redacted file keep the original's metadata, bookmarks or form fields?
No. The output is a brand-new document assembled from page images; nothing is copied across from the original except the visible appearance of the pages themselves. Titles, author metadata, annotations, attachments, form fields and layers do not carry over — which is deliberate: leftover metadata is another classic way “redacted” documents leak information.
Which quality setting should I choose?
200 DPI is the right balance for most documents: text stays crisp on screen and in print, and files stay reasonable. Choose 150 DPI when size matters most, for example for emailing, and 300 DPI for documents with very small print or fine detail that must stay sharp. Remember that at every setting the output's text is a picture — not selectable or searchable — because that is the price of a redaction that cannot be undone.