Hey PH 👋 Backstory: I spent years on KYC/AML compliance to know the importance of securing your data and checking for files metadata. Bank statements, payslips, IDs, selfies, and the question was always the same: has this file been messed with? So I built the thing I wished I had. The free tool this launch points at does exactly one job. You draw a black rectangle over a name in a PDF, hit save, send it off. Did the text actually leave the file? Very often no. It's still in there and the box is just sitting on top of it. That's how most of the famous redaction leaks happened, court filings included. Anyone who cared to select-all and copy got the "redacted" text. Tamperlens reads the page's paint order instead of hunting for redaction annotations, so it catches covered-but-not-removed text even when there is no annotation at all, which is exactly the case other checkers miss. No account, files aren't stored, and the page says plainly what it can't do (a scanned image of a covered name is beyond what byte-level parsing can prove, for instance). Underneath there's an API, the actual product, that reads revision chains, fonts, producer strings and metadata in things like bank statements, and returns evidence instead of an approve/deny verdict. We ran it over 1,728 documents nobody curated and published our false-positive rate, embarrassing part included. It's on the blog if you want to judge the engine by its worst day. What I actually want from today: tell me what's missing. A file type you needed, a check you expected and didn't find, a report that made no sense. And if you're curious what this misses, ask.

Tamperlens

Check if a PDF redaction actually removed the text

Check if a PDF redaction actually removed the text