Cover sensitive text with black boxes. Each page is saved as an image, so the text underneath is removed from the file, not just hidden.
This tool runs entirely in your browser. What you enter is never sent anywhere.Your input is never sent anywhere.
DetailsThis tool can't guarantee complete redaction. Before you share a sensitive file, open the download and try to select, copy and search the covered areas.
Why a black box in a PDF can still leak the text underneath
The usual way to hide text in a PDF is to draw a black rectangle over it. That only changes how the page looks. In most PDFs the words underneath are still stored as text, so someone can select the covered area, copy it and paste your hidden name, address or account number into a text editor.
This tool takes a different route. You mark the areas to hide, and it renders every page to an image with your black boxes painted in. It then builds a new PDF from those images. The original characters are not copied into the new file. When the PDF is ready, the tool reopens it, tries to extract text and shows you what it found.
You can draw boxes by hand, or use Find text to box every match for a name, an SSN, an email address, a phone number or a card number. Matching works on text stored in the PDF, so scanned pages need manual boxes. The automatic boxes are a starting point: always look through each page before you export.
Common uses: contracts, tax forms, medical records, FOIA releases
Contracts, invoices and bank statements
Hide names, account numbers or amounts before you send a copy to a vendor, a landlord or anyone outside the deal.
Tax forms and ID paperwork
Cover Social Security numbers, birth dates or home addresses on copies you have to hand over.
Public records and FOIA releases
Black out names or details in a document you received before you pass it along or publish it.
HR, legal and medical files
Remove personal details from case files, résumés or patient paperwork, then use the built-in text check to see whether any text is left.
How to black out text in a PDF
- Choose your PDF, or drag it onto the upload area.
- Drag on the page preview to draw black boxes. Make each box a little larger than the text. Tap a box to remove it, and use Previous, Next or the page numbers to move through the file.
- Optional: use Find text and box it to add boxes for a word, an SSN, an email address, a phone number or a card number on every page. Check each box.
- Pick a quality setting and click Create redacted PDF.
- Read the text check result, then click Download redacted PDF.
Note: Your PDF and your boxes stay in your browser while the new file is created.
Limitations and what to check before sending
- Every page is converted to an image, including pages with no boxes. Text anywhere in the file becomes unselectable and unsearchable, and the new file is usually larger.
- The new PDF is built only from page images. Bookmarks, links, form fields and attachments from the original are not carried over.
- Find text reads text that is stored as text in the PDF. A scanned page without OCR has none, so draw the boxes by hand.
- The download name is your original file name plus _redacted. Rename the file if the name itself is sensitive.
- Password-protected PDFs can't be opened. Remove the password first.
- This tool can't promise that a document meets court or agency redaction rules. Check the result yourself before you send it.
Cover-up box vs. flattened page: what stays in the file
Both look black on screen. What differs is what the file still contains.
| Black box drawn on top | This tool (page flattened) | |
|---|---|---|
| Select and copy covered text | Can still return the text | No text to select |
| Search (Ctrl+F or Cmd+F) | Can still find covered words | Finds nothing; the page is an image |
| Text layer | Stays in the file | Removed; the page is an image |
| Text outside the boxes | Stays selectable | Also becomes an image, so not selectable or searchable |
| File size | About the same | Usually larger |
| Bookmarks, links, form fields | Usually kept | Not carried over |
How the text layer works in a PDF
A short explanation of where the hidden text lives and how flattening gets rid of it.
Real characters sit under the page you see
Most PDFs made from Word, Google Docs or a web browser store text as real characters with positions. A viewer draws them on the page, and uses the same data for selecting, copying and searching.
A black rectangle is just another object drawn on top. The characters below it are unchanged, which is why a covered name can still be copied.
Scanned and OCR'd PDFs behave differently
A plain scan is a picture of a page, so there are no characters to copy or search. A scan that went through OCR usually carries an invisible text layer that matches the picture.
You can't tell which kind you have just by looking. This tool flattens every page either way, and the check at the end reports whether any text is left.
What flattening a PDF does here
The tool renders each page to a canvas with PDF.js, paints your boxes in solid black, saves the result as a JPEG or PNG, and places that image on a page of the original size in a new PDF built with pdf-lib.
Because the new file holds only images, there is no text layer to extract. The final check reopens the new PDF and counts any text it can still read.
Quality settings and file size
A page of text is tiny as text but large as pixels, so flattened PDFs grow. Small file renders at about 108 dpi as a JPEG, High quality at about 158 dpi as a higher-quality JPEG, and Lossless PNG at about 144 dpi without compression loss.
Start with Small file. Switch if small print looks too soft when you zoom in.
Redaction mistakes that expose information
Checking only how the page looks
The area is black on screen, so the file goes out without anyone trying to copy or search the covered text.
💡 See the solution▼
Using a shape or highlight tool as a redaction tool
A rectangle from a PDF editor's drawing tools covers the text visually and leaves it in the file.
💡 See the solution▼
Missing other copies of the same information
A name or number gets covered in the body, but it also appears in a header, a footer, a signature block or a later page.
💡 See the solution▼
Assuming a scanned PDF has no text
A scan that went through OCR can carry an invisible text layer, so covering the picture alone isn't enough.
💡 See the solution▼
Leaving sensitive details in the file name
The pages are clean, but the file is called something like jane-doe-ssn.pdf.
💡 See the solution▼
PDF redaction questions
Q.Why isn't a black rectangle drawn over text enough?
A. A rectangle made with an editor's shape or highlight tool only sits on top of the page. The characters underneath are still stored in the file, so you can often select the covered area, copy it and paste the hidden text into a text editor. Searching the PDF can also find the covered words.
Q.Does this tool really remove the text?
A. It converts every page to an image, paints your black boxes into that image, and builds a new PDF from the images. The original characters are not copied into the new file. After export, the tool reopens the new PDF, tries to extract text from it and reports how many characters it found. Even so, open the downloaded file and try to select, copy and search the covered areas before you share it.
Q.How do I redact every copy of a name, SSN or email address?
A. Use Find text and box it. Type a word or phrase, or pick SSN, Email, Phone (US) or Card number. The tool reads the text on every page and draws a box over each text chunk that matches. A chunk can be wider than the match itself, and a match split across two chunks can be missed, so look through every page before you export. SSN matching expects the 123-45-6789 format with dashes.
Q.Will the PDF look different afterward?
A. A little. Every page, including text outside your boxes, becomes an image. Small file renders pages at about 108 dpi, Lossless PNG at about 144 dpi and High quality at about 158 dpi. Small text may look softer when you zoom in. Choose High quality or Lossless PNG if sharpness matters.
Q.Why is the redacted PDF larger than the original?
A. Text is stored as compact data, while an image stores pixels for the whole page. Text-heavy PDFs therefore tend to grow. The result panel shows the original and new sizes side by side. Try Small file first; Lossless PNG usually produces the largest file.
Q.Does it work on scanned PDFs?
A. Yes. A scan is already a picture, so text search finds nothing and you draw the boxes by hand. If the scan went through OCR, it may carry hidden text that matches the picture. Flattening the pages removes that hidden text too.
Q.Can I redact some pages and leave the rest as searchable text?
A. No. Every page is converted to an image, including pages with no boxes, so the whole new file loses selectable and searchable text.
Q.What is lost in the new PDF?
A. The new file contains only page images, each at the original page size. Bookmarks, links, form fields, attachments and the original document properties are not copied over.
Q.Can I open a password-protected PDF?
A. No. This version can't open PDFs that need a password to view. Remove the password in another app first.
Q.Is my PDF uploaded anywhere?
A. No. Reading the file, drawing boxes, flattening pages and creating the new PDF all happen in your browser, and nothing is sent to our servers. When the tool loads, your browser downloads the PDF.js worker script from a public CDN (cdnjs). That is code coming to your browser; your file does not go out.
Q.Can I redact a PDF on a phone or tablet?
A. You can draw boxes by dragging with a finger, and tapping a box removes it. Every page is rendered as an image in memory, so a very long PDF may be slow on a phone. If that happens, try a shorter file.
Q.Is this redaction accepted by courts or agencies?
A. We can't say. Requirements differ by court and agency, and this tool doesn't certify anything. Check the rules that apply to your filing and verify the output yourself.
Your feedback
makes this tool better.
Tell us which features you'd like, what should be fixed, or which new tool you'd like to see. An engineer reads every submission and uses it to guide development.