Drop. Read. Clean. Verified.
Built for people who aren't metadata experts — powered by an engine that very much is.
Drop a document
Drag any PDF or Word (.docx) file onto the window, or browse for it. The scan runs entirely on your PC and finishes in seconds. Nothing is uploaded — there is nothing to upload to.
Read the report
You get a privacy score from 0–100 and a list of findings in plain language, grouped by category and ranked by severity. Expand any finding for the exact technical details — field names, URLs, coordinates, recoverable text.
Clean & verify
Toggle anything you want to keep, then create the cleaned copy. The new file is automatically re-scanned; you see the verified before/after score and exactly where the clean copy was saved.
The privacy score
One number a non-expert can act on. Every finding subtracts from 100 based on severity.
Active tracking: beacons, auto-run actions, remote templates. A document that phones home is never “mostly fine”.
Identifies you or your machine: author names, GPS data, recoverable deletions, leaked paths.
Circumstantial: timestamps, software fingerprints, revision counters, unique IDs.
Nothing found. A cleaned copy only gets this grade by passing a fresh scan — not by assumption.
Under the hood — and why it matters
Zero third-party parsing code
The engine reads OOXML, PDF, JPEG/EXIF and PNG with its own parsers, written from the file-format specifications — no external libraries, no supply chain, nothing opaque. For a tool you trust with sensitive documents, being fully auditable isn't a detail; it's the point.
No networking code exists
Most apps promise not to send your data. PDF Doc Shield can't: the application contains no HTTP client, no sockets, no update pinger, no crash reporter. Verify it yourself with any firewall — the app never makes a single connection.
Whole-file analysis, not just the metadata fields
The PDF scanner parses every object in the file — including compressed object streams — rather than trusting the file's own index, which is exactly where leftovers hide. The Word scanner walks every part of the package: document, headers, footers, settings, relationships, comments, embedded media.
Rebuild, don't erase
Deleting metadata in place can leave recoverable traces. Cleaned PDFs are re-serialized from scratch: only the current, visible document graph is written to the new file. What isn't written can't be recovered.