Redact PDF
Redact a PDF so the text is actually gone.
Draw a box over anything sensitive. The file you download is built from pictures of the pages with the boxes painted in, so there is nothing underneath to copy. It all happens in this tab.
- Your PDF is never uploaded
- Text under boxes is removed, not covered
- The new file is read back to check
- Free for 3 pages
PDF redaction tool
Drop a PDF to redact
The file is opened inside this page and goes nowhere else. Draw boxes over what must go. The new PDF is rebuilt from page images, so the text under the boxes is not in it.
Add another PDF (Pro)No upload. This page's security policy forbids network connections, so your file cannot leave it. See how to check that yourself.
Free: manual boxes, up to 3 pages per file. Pro ($19, once): unlimited pages, auto-find, batch redaction. How it works: each page is drawn to an image, the boxes are painted into the image, and the output PDF is built from those images alone. The original text, fonts, annotations and properties are not copied.
How it works
Open the PDF
Choose a file or drop it on the box above. Pages appear as pictures you can draw on.
Draw boxes, or find text
Drag to draw a box. With Pro, tick what to look for, press Find, and accept or reject each item in the list.
Redact and download
The page builds the new PDF, reads it back to confirm it holds no text, and saves it. Then run the checks below.
How the redaction works, honestly
Most "redact" buttons in PDF software can be done properly or improperly, and a file that merely has a black shape on top looks identical to one that is truly redacted. So here is exactly what this tool does, in order:
- Your PDF is opened with pdf.js, Mozilla's open-source PDF reader, bundled with this site.
- Each page is drawn onto an image in your browser at the quality you choose.
- Your boxes are painted onto that image as solid black.
- The image becomes a JPEG, and a small PDF writer in this site's code builds the output from those JPEGs. The writer can only place pictures. It has no way to copy text, fonts, links, comments, form fields, attachments or document properties from your original.
- The finished file is opened again with pdf.js and the page reports what it can extract. A correct result says 0 characters, 0 annotations, 0 attachments.
The automated tests for this site check the same thing against real PDFs: the redacted strings are absent from the extracted text and from the raw bytes of the output, and glyph pixels under a box are black for several fonts, sizes and rotated pages. They test synthetic documents, so treat them as evidence about the method, not a guarantee about every PDF in the world.
What you give up
- Searchable text. The whole output is images. If you need text, keep the original privately and share the redacted copy.
- Some crispness. Pages are rasterised. At the Sharper setting small print stays readable; at Small file it is softer.
- Links, bookmarks and form fields. They are not carried over, which is the point, but also means the output is not interactive.
What auto-find (Pro) finds, and what it does not
| Kind | Matches | Does not match |
|---|---|---|
| US SSN | 123-45-6789 and 123 45 6789 (also ITIN-style numbers starting with 9) | Nine digits with no separators, partial numbers |
| Card-like | 13 to 19 digits, grouped by spaces or hyphens, that pass the Luhn check | Numbers that fail the check, such as a typo |
| name@domain.tld, including plus tags and subdomains | Addresses split across lines | |
| Phone | (555) 010-0199, 555-010-0199, 555.010.0199, +1 555 010 0199 | Ten digits with no separators, non-US formats |
| Dates | 03/15/2025, 2025-03-15, March 5, 2025, 5 March 2025 | Dates written in words with no year |
| Your terms | Literal terms (any case) and /regex/ lines | Anything you do not type |
False positives cost you a click in the review list. False negatives cost more, so the finders lean towards finding too much. Boxes for found text are placed from the text positions in the PDF, with a margin, and the preview shows exactly what will be blacked out.
Verify your redaction
Never trust a redacted PDF because it looks right. The classic failure is a black rectangle drawn on top of text that is still in the file: anyone can select, copy or search the text underneath. Run these checks on any redacted PDF, whichever tool made it, before you send it.
- Search for it. Open the redacted file, press Ctrl+F (Cmd+F on a Mac) and search for a word you redacted. A real redaction finds nothing. Search for a nearby word you did not redact: if that is found, the file still has a text layer, so check what else is in it.
- Select and copy. Press Ctrl+A, copy, and paste into a plain text editor. Look for the redacted text in what you pasted.
- Drag the box. In an editor that lets you move objects, try to move or delete the black rectangle. If text appears under it, the redaction was only a cover.
- Read the text with a tool. On a command line,
pdftotext file.pdf - | grep -i "the secret"prints the text layer of every page. Nothing should match. - Check the extras. Open the document properties (title, author, subject, keywords) and look at the comments, attachments, bookmarks and form fields panels. Redaction of the pages does not clean these.
- Look at every page. A box that is a few pixels too small leaves the edge of a letter visible, and an unredacted second copy of a number elsewhere on the page is easy to miss.
Plainsum PDF's output passes checks 1, 2, 4 and 5 by construction: the new file is built from page images only. After each redaction the page also re-opens the new file and reports how many characters of text, annotations and attachments it found (it should say zero). Checks 3 and 6 are yours to do.
Questions
Is my PDF uploaded anywhere?
No. The file is opened by code running in your browser tab, and the page's Content-Security-Policy forbids network connections. The Network tab in your browser's developer tools stays empty while you redact.
Is the text really removed, or just covered by a black box?
Removed. Each page is drawn to an image inside your browser, the black boxes are painted into that image, and the output PDF is built from those images only. Nothing from the original file is copied across, so there is no text, font, annotation, form field, attachment or document property for the boxes to leave behind. After building, the page opens the new file again and reports the number of text characters it can extract, which should be zero.
Why can I not select text in the redacted PDF?
Because every page is now a picture. That is the price of guaranteeing the hidden text is gone, and it applies to the parts you did not redact as well. Keep the original if you need searchable text, and share only the redacted copy.
What does auto-find look for?
Pro lists US Social Security numbers (NNN-NN-NNNN), card-like numbers of 13 to 19 digits that pass the Luhn check, email addresses, US-format phone numbers, dates, and your own terms or regular expressions. It does not find names, addresses or free-form account numbers. Every hit is shown for you to accept or reject, and you should still look at every page.
Can it redact a scanned PDF?
You can draw boxes on a scan, because a scan is already a picture and the box is painted into it. Auto-find cannot read a scan, because there is no text and this version has no OCR.
What changes about the file?
The page size and page order stay the same. The pages become JPEG images at 110, 150 or 220 dpi (you choose), so text is a little less crisp when you zoom in and the file size can go up or down. Document properties such as title and author are not copied.
What is free and what is Pro?
Free: manual boxes on up to 3 pages per file. Pro, one payment of $19 that also covers the compare tool: unlimited pages, auto-find, and batch redaction of several files into one zip.
Is this legally enough to release a document?
It is a tool, not legal advice, and it cannot decide what in your document is sensitive. It guarantees one specific thing: text under a box is not in the output file. Check every page of the result yourself before sharing it.