Permanently black out sensitive text, IDs, bank details and signatures — drawn by you, destroyed for real. Every page is re-rendered in your browser with the boxes burned into the pixels, so the hidden content cannot be copied, selected or recovered. Your un-redacted original never leaves your device.
Contracts, statements, medical records, ID copies — one PDF at a time
This PDF has 0 page(s). Drag on a page to draw black boxes — 0 box(es) drawn in total.
Search for "redact pdf online free" and the big names appear: Adobe's own web tool, Smallpdf, iLovePDF, Sejda. Nearly all of them work the same way — you upload your document to their server, the redaction happens there, and you download the result. For a public flyer that is harmless. For the documents people actually redact, it is the worst possible architecture.
Think about what gets redacted: tax returns with Social Security and national ID numbers, bank statements with account numbers and balances, medical records with diagnoses, court filings with minors' names, contracts with pricing and signatures, HR files with salaries and home addresses. These are the most sensitive documents most people own — and the standard "free" redaction flow asks you to hand the un-redacted original to a stranger's server first. Their privacy pages promise deletion after an hour or two, but a promise is not architecture. While your file sits on their disk it can be copied, breached, subpoenaed or mined, and you will never know. The strongest privacy guarantee is not a policy — it is a design where your file never leaves your device in the first place.
Then there is the second, quieter problem: most "redaction" on the web is not redaction at all. Plenty of tools — and plenty of do-it-yourself advice — simply draw black rectangles on top of the text. The words are still in the file, fully selectable, fully extractable, recoverable in seconds by anyone who selects the text or opens the document's text layer. High-profile leaks have happened exactly this way: a black box that looked safe, peeled away to reveal everything underneath. A tool that only covers text is a highlighter with bad intentions.
This page was built to close both gaps at once. The redaction happens entirely in your browser — your un-redacted file is never uploaded anywhere. And it is genuine destruction, not decoration: every page is re-rendered as a fresh image with solid black pixels burned in where your boxes were. The original characters are gone from the output file, not hidden behind it. You can verify both claims yourself in under a minute: watch your browser's network tab while redacting (zero uploads), then run the select-and-copy test on the exported file (nothing to select).
A PDF is not a photograph of a page — it is a set of instructions: draw these characters here, at this position, in this font. When a basic editor draws a black box "over" sensitive text, it only adds one more instruction: draw a black rectangle here. The original instruction — the actual text — stays in the file, underneath, perfectly intact. Anyone who selects the region, extracts the text layer, or simply deletes the rectangle layer gets the secret back. That is why "blacked out with a marker tool" is the single most common redaction failure in the wild.
True redaction removes the content instead of covering it. Professional tools do this by deleting the text operations from the page's content stream. This tool takes a simpler route with the same guarantee: it re-renders each page into a plain image and burns your black boxes into the pixels before rebuilding the PDF. In the exported file there is no text instruction left to recover — the page is just dots, and where your box was, the dots are black. The trade-off is honest and worth stating plainly: because the pages become images, the output PDF has no selectable or searchable text layer. If you need a searchable document afterwards, keep your original — the tool never modifies it, and always writes a separate "-redacted" copy.
One more thing the black-box method cannot fix, and neither can pixels alone: metadata. Document properties — author name, title, the software that created the file, sometimes the full revision trail — can leak identity even from a perfectly redacted page. That is why the export pipeline also strips document metadata as it rebuilds the file. For maximum hygiene, run the result through our metadata remover as a second pass before sharing, and if the file was password-protected to begin with, unlock it here first so you can open and redact it locally.
Redaction is rarely a one-step job. A sensitive document usually travels a small pipeline: open it, hide the secrets, strip the invisible data, shrink it for sending, and combine it with related pages. Because every tool on this site runs in your browser with the same no-upload design, you can chain them without your file ever touching a server. Here are the workflows people actually use:
Every one of these tools is free, unlimited and private by the same architecture — the hub below links all eighteen, so you never have to leave this workflow or upload your document anywhere.
Permanently black out sensitive text — burned into the pixels, never uploaded.
Open tool →Remove the password from a protected PDF — decrypted in your browser, never uploaded.
Open tool →Strip hidden author/title/software data — see it first, 100% in your browser.
Open tool →Lock filled forms permanently — freeze fields into the page, 100% in your browser.
Open tool →Draw your signature and stamp it onto any page — no printing, no upload, no account.
Open tool →Most people under-redact. They black out the obvious account number and miss the rest. Before exporting, walk the whole document — headers, footers, margins and all — and check for these:
Common scenarios: a job applicant sharing a previous payslip with the salary blacked out but the address left visible; a freelancer sending a client contract with old pricing redacted; a seller sharing a vehicle document with the VIN hidden; a patient posting a lab report with name and record number removed; a lawyer filing evidence with a minor's name blacked out. In every case the rule is the same: if a stranger reading the document should not know it, it gets a box. When in doubt, redact — you can always share the un-redacted original later, but you can never un-share a leak.
Adobe, Smallpdf and iLovePDF redact your file on their servers — meaning the un-redacted original travels to a stranger first. Here the entire job — rendering, box drawing, re-export — runs in JavaScript on your device. For documents defined by their sensitivity, that architecture is the whole point.
Many free tools just paint black rectangles over your text, leaving the words fully recoverable underneath. This tool re-renders every page as an image with the boxes burned into the pixels — the original characters are gone from the output file, verified by the select-and-copy test.
A perfectly blacked-out page can still leak your name through document properties. The export pipeline clears title, author, subject and keywords as it rebuilds the file — and our metadata remover gives the result a second pass for belt and braces.
Server tools ration free redactions because servers cost money. There is no server here, so there is nothing to ration: redact as many documents as you need, whenever you need, with no sign-up and no "3 free tasks per day" wall.
Drop your PDF on the tool above, drag black boxes over everything you want hidden, and press "Export redacted PDF". Your browser re-renders each page with the boxes burned into the pixels, so the content is permanently destroyed — no Adobe, no account, no upload, no cost.
No. Opening the PDF, rendering the pages, drawing the boxes and building the redacted file all happen in JavaScript inside your own browser. The un-redacted original never leaves your device — the safest possible setup for sensitive documents.
No. This tool does not draw a layer over your text — it re-renders each page as a new image with solid black pixels where your boxes were. The original characters no longer exist in the output file, so there is nothing to select, copy or recover with any PDF tool.
A black highlighter only covers text visually: the words underneath stay in the file and can be copied or extracted. Redaction destroys the content — after exporting here, the hidden text is gone from the file entirely. They look similar on screen but are completely different in safety.
No — and that is by design. Because every page is re-rendered as an image to guarantee the redacted content is destroyed, the output has no text layer, so nothing is selectable or searchable. Layout and appearance are preserved at high resolution.
Yes — scanned PDFs redact even more reliably, because the text is already just pixels. Draw boxes over the scanned words, export, and you get a clean image PDF with those areas permanently blacked out.
No artificial limits. Everything runs on your own device, so there is no server to ration — redact a 2-page letter or a 200-page report. Very large documents simply take longer to re-render on your machine.
No — redaction here is permanent, which is exactly what makes it trustworthy. Always keep a copy of your original. The exported file is named after your original plus "-redacted", and your original is never modified.
Permanently black out sensitive text — burned into the pixels, never uploaded.
Open tool →Remove the password from a protected PDF — decrypted in your browser, never uploaded.
Open tool →Strip hidden author/title/software data — see it first, 100% in your browser.
Open tool →Lock filled forms permanently — freeze fields into the page, 100% in your browser.
Open tool →Draw your signature and stamp it onto any page — no printing, no upload, no account.
Open tool →