How to check a PDF for hidden metadata without uploading it
A PDF is a container, not just a page. Beyond the visible text and images it stores structured data most people never see: who wrote it, with what software, when it was created and changed, what was edited along the way, files tucked inside it that never appear on any page, comments left by reviewers, and layers that are switched off. That data travels with the file to whoever you send it to — a client, a publisher, a court, the public — and the usual advice is to run ExifTool on the command line or dig through Acrobat's Properties, Attachments and Sanitize dialogs, none of which gives you a verdict, a batch check, or a report you can send to the person who made the file.
This tool does exactly that, where the file already is: on your own device. When you drop a PDF here, your browser:
- Checks the file is really a PDF — by its %PDF magic bytes, not its
name — and that pdf.js can open it, so a corrupt or truncated file is refused with a
clear message instead of producing a wrong report. A JPEG renamed to
.pdfis refused the same way, and a PDF renamed to.txtstill inspects. - Reads every metadata store the file carries — the document information dictionary, the XMP stream parsed field by field (including edit history and document IDs), embedded files and attachments, annotation metadata, optional-content layers, incremental-update revision history, and JavaScript or automatic actions — including metadata stored in compressed object streams, which most modern producers use. Embedded images are counted, because photos in a PDF often keep their own EXIF data.
- Gives each file a verdict — Clean, Review before sharing, or Exposed — with every finding labelled by what it reveals and how to remove it, so you know what to fix and where. A downloadable text report per file (or one combined report for the batch) carries the same detail to whoever needs it.
What makes this inspector different
The pages that top the results for this search mostly check a single file on a server, charge for the detailed report, or need an account — and the one dedicated client-side rival we found shows only creator information and dates for a PDF, with no XMP parse, no embedded files, no annotations, no revision history, no verdict and no downloadable report. This page is the only fully client-side dedicated PDF hidden-metadata inspector we found that combines a batch of 20 PDFs with a per-item risk verdict and removal fix, a full XMP parse with edit history, embedded-file and annotation listings, hidden-layer and revision detection, and a free downloadable report for every file — with no daily limits, no account and no watermark — plus a live network-request counter so you can verify the privacy claim yourself. It runs entirely in your browser — there is no server to upload to.
When is inspecting the right choice — and when is it the wrong one?
Inspecting is right whenever a PDF leaves your hands: a manuscript for a publisher, a contract or proposal for a client, a report for a journalist or a court, a CV, a book for KDP or IngramSpark, or anything posted publicly. It is the wrong choice for a file you only ever open yourself, where the metadata never leaves your machine — and for a file whose metadata you actually want to keep, such as an archived document where the author and dates are part of the record. If the destination accepts the source document, fix the metadata there instead of patching the PDF.
Honest limitations
Metadata, not full forensics
This tool reads metadata — it is not a forensic engine. It cannot recover text deleted in an earlier revision, only tell you revisions exist.
Embedded image EXIF is not read
Photos inside a PDF often keep their own EXIF data — camera model, GPS, timestamps. This tool counts the images and says so; it does not open the image bytes.
No guarantee of cleanliness
A producer can hide data in ways no inspector can enumerate. A Clean verdict means nothing this tool can see was found — not that nothing is there.
Some metadata is invisible
Metadata inside encrypted streams, or stored by an application in a private structure this tool does not read, will not be listed.
No password-protected PDFs
Password-protected PDFs are refused, and this tool never asks for a password. Remove the password in a desktop PDF tool first, then inspect the unlocked copy.
Scanned PDFs still carry metadata
A scanned PDF has no fonts or text, but it still carries document properties, XMP and the scanner's software trail — and its embedded page images may carry EXIF. Check it the same way.