How is this different from drawing a black rectangle?

A black rectangle is a shape sitting on top of the text. The words are still in the file, and anyone can select them, copy them, or extract them with a command-line tool in seconds. This rebuilds the page from pixels instead, so there is no text object left to recover — because there is no text on the page at all.

Does that mean the whole document becomes an image?

No, and this is the part worth understanding. Only pages you actually mark are rebuilt. A forty-page contract with one redacted page keeps thirty-nine pages of real, selectable, searchable text. Pages you do not touch are copied across untouched.

What do I lose on a page I redact?

All of its text stops being selectable and searchable, not only the part you blacked out, and a screen reader can no longer read it. That is the price of the guarantee. There is no way to remove text permanently while leaving the rest of the page as live text without rewriting the page content stream, and a tool that claims to do that in a browser is worth distrusting.

How do I check it actually worked?

Open the saved file, try to select the text where a black box is, and paste it somewhere. You should get nothing. Ten seconds of checking is worth more than any promise on this page, including ours.

Is anything left behind?

The page content is gone. Document metadata, attachments and bookmarks are not part of the page and are not removed by this — if a name appears in the document title or the author field, it is still there. That is a genuine limitation, not a detail.

Redact a PDF — Text Deleted, Not Covered

Black out text so it is genuinely deleted, not just hidden under a rectangle. The page is rebuilt from pixels, so nothing can be selected or recovered.

The problem with almost every free redaction tool

They draw a black rectangle over the words.

The rectangle is a shape. The text underneath is still in the file, exactly where it was. Anyone can select it and copy it, and anyone with a command-line tool can extract the whole page in about a second. The information is not hidden; it is behind a picture of a black box.

This is not a hypothetical. It has leaked real names out of real court filings and real government releases, repeatedly, and it keeps happening because the tool looked like it worked.

What this does instead

A page you redact is rebuilt from pixels.

The page is rendered to an image, the black boxes are painted onto that image, and the image replaces the page. Afterwards there is no text object anywhere on that page to recover, because there is no text on the page at all.

That is the only way to be certain in a browser, and it is why the promise here can be checked rather than believed.

Only the pages you mark

Rasterising an entire document to redact one line would be a bad trade, so that is not what happens.

Pages you mark are rebuilt. Pages you do not mark are copied across untouched, and keep their real, selectable, searchable text. A forty-page contract with one redacted page comes back as thirty-nine ordinary pages and one image.

What a redacted page costs you

Stated plainly, because it is real:

All of that page's text stops being selectable and searchable — not only the part you blacked out.
A screen reader can no longer read that page.
The page will usually be somewhat larger, because a picture of text costs more to store than text.

There is no way around this. Removing text permanently while leaving the rest of the page as live text means rewriting the page's content stream, and any browser tool claiming to do that reliably is worth distrusting.

Check it yourself — do not take our word for it

Open the saved file. Try to select the text where a black box is. Paste it somewhere.

You should get nothing. That check takes ten seconds and is worth more than any claim on this page, including this one.

For what it is worth, we run the same check automatically: the test suite extracts the raw text back out of a redacted file and fails if the secret string survives, with a control test proving the extractor works — so a broken extractor cannot make the result look good.

What is still left behind

The page content is gone. These are not:

Document metadata — title, author, keywords. If a name is in the document properties, it is still there.
Attachments and bookmarks — not part of the page, not removed here.
The original file on your own machine, which is a good thing until you send the wrong one.

If you are producing something for a court, a regulator or a journalist, treat this as one step rather than the whole job.

Why this has to run in your browser

The documents people redact are the ones they least want on someone else's server: medical records, contracts, statements, filings. Uploading them to a stranger to have the sensitive parts removed is a strange thing to do, when you think about it for a second.

Nothing here is uploaded. Watch the Network tab if you would like to confirm it.

This tool runs entirely inside your browser. Your file is never uploaded — it never leaves your device, so there is nothing for us or anyone else to read, store or leak.

Permanently redact a PDF