A PDF you are about to send often knows more than the page shows: who wrote it, which company owns it, what a colleague wrote in the margin, an earlier draft still embedded inside, an invisible watermark tracking the copy. Most privacy scanners hand you a report and stop there. This one closes the loop.
Basic scanning runs locally in your browser. No upload required for standard checks.
"Hidden data" in a PDF is not one thing. It is six different places a document can remember, and they survive saving, converting and emailing.
The document properties hold the author name, the organisation field, the creating application and full creation/modification timestamps. A contract sent to a counterparty can reveal your internal file paths and who drafted it.
Annotations and comments are stored as separate objects. People routinely strip the visible text and forget the review thread — which is exactly the part that says "do we really want to commit to this clause?"
Text rendered in the background colour, or positioned outside the visible page box, is invisible to the eye but still selectable, searchable and extractable by any parser.
A PDF can contain attachments and previous versions as embedded objects. You send the final file; the recipient can open the version you decided against.
Some watermarks are drawn as content-stream text objects you can remove. Others are forensic or "dark" markings — low-contrast, per-copy identifiers embedded for traceability. The scan distinguishes them.
Text that was struck out or deleted during editing frequently remains in the structure underneath. Visually gone, technically still there.
This is the part most scanners leave to you. A report that says "this file contains 14 pieces of hidden data" is useful for about ten seconds — until you need to do something about it.
You get a list. Then you open a second tool to remove metadata, a third to redact text, a fourth to add your own annotation, and a fifth to convert the file. Every hand-off is another upload, another copy left on another server, and another chance to send the wrong version.
Scan, see exactly what was found, remove it at object level, annotate what needs explaining, and export a clean file with a record of what was done. The document never leaves the workspace between steps.
Upload or drop the file. Standard checks run in your browser; the scan reports metadata, comments, hidden objects, embedded files and watermarks.
Delete what should not travel — permanently, at object level, not by painting a black rectangle over the text.
Add the notes that do need to travel: review comments, approval marks, redaction reasons.
Export to PDF or another format, then re-parse the result to confirm the removals actually held.
Four minutes, and it catches the things people get fired over.
No. Metadata is one of six places a PDF can hold information. A scan also covers comments, invisible text, embedded files, revision remnants and watermarks. Removing metadata alone leaves the other five exposed.
Standard checks — metadata, comments, hidden objects, embedded files — run locally in your browser. Only features that genuinely need server-side processing (such as AI-based image watermark removal) send the file, and that is stated before you confirm.
A low-contrast or per-copy marking embedded for traceability rather than branding. It is often invisible on screen but recoverable from the file structure. Detecting it requires inspecting the content stream and object tree, not looking at the rendered page.
Because the text stays in the file. Anyone can select it, copy it or extract it. Proper redaction removes the underlying text object so nothing remains to recover.
Yes, and it is worth doing. Documents received from counterparties carry their own metadata, and files that reach you through a chain often still contain earlier drafts.
PDF as the primary output, with conversion to Word, Excel, image and plain text where you need to hand off an editable or flattened version.
Scan, redact, annotate and export in one place — and verify the result instead of assuming it.
Start with a free scan