Sometimes you do not want to change anything — you want to know what is in the file before it leaves your computer. Drop a PDF here and read the list, then decide.
The tool runs in your browser (JavaScript required). Nothing is uploaded.
The document is read twice, by two independent readers, and the text is examined wherever it is stored — on the page, in comments, in form values, in bookmarks, in the properties and in the XMP packet. Personal information is found by the same detectors used for photos and Word documents, and you can add your own terms.
The honest part of a privacy tool is what it admits it did not do. There is no OCR here, so text inside a scan or a photograph is not read and the page is reported as not fully checked. Attachments are named but not opened. A signature is detected, not validated. An encrypted PDF is inspected, but no copy is made of it.
Whatever you decide afterwards, the copy is judged on its own bytes: reopened, searched and rescanned, with the result showing what was found before next to what the second look found.
No. The list appears as soon as the file is read. Making a copy is a separate step that you start yourself.
No. It is read inside this browser tab, and nothing about its contents leaves the page.
That part of the file could not be read well enough to make a claim about it — a scanned page, an unreadable font encoding, a parse problem. The result stays at that level instead of being reported as clean.
Only if it opens without a password. A PDF that needs one to open cannot be read at all, and it is reported as not checked rather than as empty.
Up to 50 MB and 1000 pages in this version. Larger files, and files built to unpack into far more data than their size suggests, are refused with a message instead of being processed.