pdfpix

Check a PDF for redaction leaks

or drop one here

What this check can find

Opaque dark rectangles with selectable text underneath, plus document metadata, annotation text, attachments and hidden layers that need review.

It does not certify a PDF as safe. Unusual covers and words stored only inside images can require a manual review.

Choose a PDF first.

A black box is not proof that the text underneath was removed. This check reads the PDF drawing instructions, outlines supported cover boxes and reports any selectable text still below them. It also lists private data outside the page. The PDF never leaves this tab and is not changed.

How to check a PDF redaction

  1. Choose a PDF

    Press Choose a PDF, or drop one on the page. The file stays in this browser tab.

  2. Inspect every page

    The checker reads page drawing instructions, finds opaque dark rectangles and compares each one with the selectable text layer below it.

  3. Review recovered text

    Pages with cover boxes appear in the report. A red outline marks any box with live text underneath and shows the recovered text in full.

  4. Check the limits

    Read the metadata, annotation and hidden-layer findings, then review the stated limits. No leak found is not a safety certificate.

What this result means

A PDF page is a stack of drawing instructions. A failed redaction often leaves the original text in that stack and adds a dark rectangle above it. The page looks clean, but copying the text layer or reading the content stream reveals the hidden words. This tool looks for that specific failure and shows the text it can recover.

The result has a hard boundary. A black photograph, an unusual clipping path or a cover made from many small shapes may not look like one vector rectangle to the parser. Text burned into an image has no selectable text layer to compare. A result that says no leak found means these checks found none. It does not mean every possible redaction technique was proved safe.

What a redacted page should look like

A properly redacted page has nothing underneath the black. The characters are gone from the content stream, so selecting the area copies nothing and searching for the name finds nothing. A page that only looks redacted still holds every character below the rectangle, and any reader can lift it out. Redact PDF removes the text rather than covering it, and Remove PDF metadata clears the copy of a name that may still sit in the document properties after the page itself is clean.

Questions

Is the checked PDF uploaded?
No. The PDF, page previews and recovered text stay in this browser tab. Analytics receive aggregate counts only.
What is a fake redaction?
It is a dark shape drawn over text without removing the text below it. The page looks covered, but the text can still be selected or extracted.
What cover boxes can this check find?
It checks opaque dark vector rectangles in the PDF page content, including rectangles drawn inside reusable form objects.
Does it inspect metadata and annotations?
Yes. It lists supported document metadata, attachments, annotation text and optional-content layer groups for manual review.
Does no leak found mean the PDF is safe?
No. It means these checks found no selectable text below the supported cover boxes. Unusual cover techniques and text stored only inside images still need manual review.
Does this tool change my PDF?
No. It produces a local report and never writes a new file. Use Redact PDF to permanently remove marked content.

The other pdfpix tools

All 30 of them run in a browser tab like this one. Each reads its own output back and checks it before offering the download, and none of them needs an account.

Where to go next