Redact PDF
Finds every occurrence of the words you name and destroys them, rather than drawing a box over them. Pages containing a redaction are rebuilt as images, so there is nothing left underneath to recover.
Local processingAccepts PDFUp to 100 MB
Drop your file here
PDF · up to 100 MB each · you can also paste from the clipboard
Nothing is processed until you press the button below.
How this tool works
- 1The text layer is searched for the terms you give, and every match is located on the page.
- 2Each page holding a match is rendered to an image, the matched areas are painted over, and the page is replaced by that image.
- 3Pages with no matches are copied across untouched and keep their text layer.
What it will not do
- A redacted page loses its entire text layer, not just the redacted part. That is the cost of the removal being real, and it is why only the affected pages are rebuilt.
- Only text that exists as text can be searched for. Words that are part of a scanned image are invisible to the search and have to be marked by hand.
- Redacted pages become images, so the file will usually be larger.
- Check the result. Extract the text from the output and search it yourself before sending it anywhere.
Common uses
- Removing a name or address before disclosing a document.
- Stripping a reference number from a contract being shared as a template.
Where this runs
Redact PDF runs entirely inside this page. Your file is read by the browser, processed in memory and handed back to your downloads folder. It is never sent to a server, because there is no server involved in this tool.
Read the full processing model →Questions
Because that does not remove anything. The text stays in the file underneath the box and can be selected, copied or extracted in seconds. It is the single most common way documents leak.