Redact a PDF without uploading it to any server
Browser-based redaction does the work on your own device, so the file never leaves it. Here is how that closes two risks at once, and how you can prove the text is actually gone.
You can redact a PDF without uploading it by using a browser-based tool that does the work locally, on your own device. KeptPDF runs entirely client-side, so the file never touches a server. You can disconnect from the internet and it still redacts. It permanently removes the underlying text and metadata instead of covering them, then issues a SHA-256 certificate documenting exactly what was redacted.
That last part is the whole point: a black box on the screen does not mean the words are gone, and a tool that uploads your file to “process” it has already done the one thing you were trying to avoid.
Why “without uploading” is a separate problem from “redacted correctly”
There are two different risks in handling a sensitive document, and most online tools only pretend to address one of them.
The first is the redaction itself: does the tool actually remove the text, or just draw a rectangle over it? Most generic editors only draw the rectangle. See how to redact so the text is actually gone.
The second is the transmission: to redact your file, a typical online tool uploads it to the company’s servers first. For a privileged filing, a medical record, or a client’s financials, that upload is itself a disclosure. The document now exists on someone else’s infrastructure, subject to their retention, their breaches, and their subpoenas, regardless of how technically sound the redaction that follows is. You have handed over the raw, un-redacted file just to get it redacted. (More on what that upload really means: do online PDF tools upload your file?)
Browser-based redaction closes both gaps at once. The file is read and rewritten in the page you already have open; nothing is sent anywhere to be processed.
Your raw file is uploaded to their server to be processed. The disclosure has already happened, no matter how sound the redaction that follows.
The file is read and rewritten in your browser, on your device. It is never sent anywhere to be processed, so there is nothing to disclose.
How to redact a PDF without uploading it
- Open the Redact tool. It is free and runs in your browser, with no install and no account required to start. Drop your PDF onto the page, or pick it from your device.
- Let it find the sensitive data, or mark your own. KeptPDF scans the document on your device and surfaces names, SSNs and other government IDs, account and routing numbers, emails, phone numbers, addresses, and dates for you to review. You can also draw your own black boxes over anything it did not flag.
- Review every match. Auto-detection is a first pass, not a guarantee of completeness, so you confirm each one. When you confirm a term, every exact match of it is removed across the whole file, which matters in a long document where the same name appears 200 times.
- Apply. KeptPDF destroys the text and image content under each mark, scrubs the document’s hidden metadata, and flattens the result so no hidden layer survives.
- Download the redacted PDF and its certificate. The file is processed entirely on your device and was never uploaded at any step.
How you know the text is actually gone
This is the part that separates a real redaction from a black box, and it is where the skeptical reader should be paying attention.
When KeptPDF finishes, it re-opens the file it just produced and confirms that zero text remains under the marked areas. It does not take its own word for it. It then computes two fingerprints: the SHA-256 of the original file as you provided it, and the SHA-256 of the redacted file you download. Both are written into a certificate, alongside a plain-English log of what was removed.
The certificate also carries a verification link. Anyone you send the redacted file to (opposing counsel, a court clerk, a client) can open that link, drop the PDF in, and confirm its SHA-256 matches the certificate. The file they drop is hashed in their own browser and never uploaded, so the check itself leaks nothing. It is a tamper-evident record you can produce if your process is ever questioned. It is not a promise about how a court will rule, just proof of exactly what you removed and that the file has not changed since.
What’s free and what’s paid
Redacting a single PDF (true text removal, metadata scrubbed, processed locally, nothing uploaded) is free. Free output carries a small KeptPDF watermark, and you can see a locked preview of your own certificate.
A paid plan (Pro) removes the watermark and unlocks the downloadable proof certificate: a branded PDF, plus JSON for your records. For firms, the practice plan adds batch redaction across an entire production and shared team redaction lists, so a name or a privilege term is removed consistently across every file in a set, with an audit log of every removal.