0 requests since your file/works offlineYour desk

Is that redaction real?

Drop a redacted PDF. Each page is rendered and compared with its text layer to find words that sit under dark boxes, which means they were covered, not removed.

Diagnostic · runs in this tab · free
Redaction checkerRead in this tab
Drop a redacted PDFEach page is rendered here and compared with its text layer.

The checker flags text whose area is almost entirely dark on the rendered page. It cannot see text that exists only as an image.

Results appear here: text hidden under boxes by page, redaction marks that were never applied, and metadata that may name the author.

Requests since you added a file: 0Counted by this page · check it without trusting us

How the check works

For every piece of text in the PDF’s text layer, the checker looks at the rendered page under that text. If the area is almost entirely dark, the text is drawn but covered: it is still in the file and anyone can copy it. Those are listed with the page number and the first characters, so you can see the problem without the whole secret being displayed.

It also reports redaction annotations that were marked but never applied, and metadata that might name the author.

It cannot find text that is only in an image (a scanned page with a box drawn on it), because there is no text layer to compare; if the image underneath was replaced, the redaction there is fine.

Steps

  1. Drop the PDF.
  2. Each page is rendered and its text positions checked against dark areas.
  3. Review the list of hidden text found, by page.
  4. Re-redact properly if needed.

Limits, stated plainly

  • Light-coloured boxes (white-out) are checked too, but text under a white box looks like empty space; the checker flags text under any opaque box it can detect.

Engine: pdf.js rendering and text in this tab; pdf-lib in a Web Worker. Your file is not uploaded; check it yourself.