🕵️ PDF Redaction Checker
Black boxes are not redaction. Before a "redacted" PDF leaves your hands, see everything a copy-paste can still pull out of it — and search for the exact name or number you meant to remove.
This checks the classic redaction failures — text hidden under drawn boxes, removable annotations, metadata and attachments. It cannot prove a negative (e.g. content encoded in images still needs human review). To redact for real, use the rasterizing redactor. Nothing you load here leaves your browser.
How the pdf redaction checker works
Most redaction failures share one mechanism: the sensitive text is still in the file, merely covered by a drawn rectangle — so selecting the area and copying, or extracting the text layer, reveals it. This checker extracts the complete machine-readable text of every page (exactly what any script or PDF library sees), tells you how many characters remain, and gives you a search box: type the name, account number or address you redacted, and it reports every page where it still exists, with context. It also flags the other classic leaks — drawn/markup annotations that any editor can delete to reveal what is underneath, document metadata (author, title, creator software, dates), and embedded file attachments that travel with the PDF unredacted.
This failure mode keeps making headlines: in December 2025, PDFs released in the Epstein files were found to have black boxes drawn over intact, selectable text, and forensic research on tens of thousands of published agency PDFs has found the majority of "redacted" documents still contained the hidden content. The check that would have caught all of it takes ten seconds and — done here — never uploads the document. One honest limitation: a checker can prove content IS still present, but cannot prove a negative; visual content inside images still needs human review.
Frequently asked questions
How do redactions fail if the page looks black?
Because appearance and content are separate layers in a PDF. A black rectangle drawn over text changes what you see, not what the file contains — the text underneath remains fully extractable by copy-paste, search or any PDF library, and if the box is an annotation, it can simply be deleted.
What does this checker examine?
Four things: the complete extractable text of every page (with a search box to hunt for the redacted content), overlay/markup annotations that could be hiding text, document metadata (author, title, software, dates), and embedded file attachments.
The checker found my redacted text — what now?
The redaction must be done destructively: either use professional redaction that removes the content from the file, or flatten the pages to images with the boxes burned in — the rasterizing redactor here does exactly that. Then re-check the result.
The checker found nothing — is my document safe?
It passed the checks that catch the overwhelming majority of real-world failures. But no tool can prove a negative: content baked into page images (like a photographed document) needs your eyes, and cropped-but-retained image data is beyond any text-level check.
Is the document I check uploaded?
No — text extraction and all analysis run in your browser, offline-capable. Documents being redacted are by definition sensitive; that is exactly why this tool has no server side.