PDFGym
EN

Scan a PDF for private data

or drop one here
Choose a PDF file first.

Open any PDF and read a grouped list of the private data riding with it: document details, software, dates, other metadata and embedded attachments. Choose what to remove, or remove everything, and download a verified copy with -cleaned in its name. The scan never touches printed page words and never sends anything anywhere.

The board pack going public

This page is the pre-send audit: open the file, read what it carries beyond its pages, and strip what strangers must not see. If the PDF goes to clients, regulators or the open web, scan it here first.

Say the comms officer must publish Board-Pack-March.pdf on Friday: forty pages of minutes and charts, exported from three different tools by two assistants. She opens the final export here on Thursday: the scan groups eleven findings, an ex-employee's name in Author, two producer tags, modification stamps, an XMP packet echoing the same, and a forgotten spreadsheet attached to page nine. She keeps the title, removes the rest, and publishes Board-Pack-March-cleaned.pdf.

Findings arrive grouped: document details such as title, author, subject and keywords; software fields naming creator and producer; creation and modification dates; other metadata with the XMP packet and the trailer identifier; and embedded file attachments with their names and sizes. Every item starts selected with its value visible, so the audit reads like an inventory, not a warning light.

Pressing Remove selected data deletes the chosen entries from the file structure and reopens the result: selected data must be gone, unticked items must remain, page count must match and selectable text must pass retention. A sweep that takes two minutes:

Four steps from unknown file to known file

  1. Choose a PDF

    Press Choose a PDF or drop the file onto the page. Locked files are refused before anything is listed; open them with Unlock PDF first.

  2. Review what it carries

    Read the grouped findings with their values. The scan covers metadata, XMP, identifiers and attachments; printed words are explicitly out of scope.

  3. Choose what to remove

    Keep everything selected for the full sweep, or untick items to keep. Select all and Deselect all reset the choice in one press.

  4. Remove and download

    Press Remove selected data. The cleaned file is reopened and checked before Download cleaned PDF appears.

Audits that miss, and why

Scanning a draft instead of the final export audits the wrong file: the published copy gains fresh stamps the moment it is exported. Always scan the exact bytes that travel, and rescan after any later touch, even a compress or a stamp.

Embedded attachments surprise people most: a spreadsheet or an earlier draft riding inside the PDF, invisible on every page. The scan names them; removing the file detaches them. Confirm the count in the proof line before sending.

Printed names are outside this tool's reach by design. A letterhead, a footer credit or a signature image survives every metadata sweep, so pair the scan with Redact PDF when pages carry their own identifiers.

When a scan is not enough

A scan reports and removes metadata; it never judges page content. When the words on the pages identify people, Redact PDF destroys them first, then this scan cleans the envelope, in that order.

Remove PDF metadata runs this identical sweep under a narrower name, ending in -metadata-free instead of -cleaned. Use either door. When the cleaned file must also stay closed, Password protect PDF locks it next.

Short answers about privacy scans

How is this different from Remove PDF metadata?
Only the name and the output suffix: same engine, same findings, same checks. This page downloads -cleaned files; that page downloads -metadata-free ones.
Does the scan read my page text?
Only to verify it survives: the retention check compares selectable text before and after. No printed word is listed as a finding or removed.
Can I scan without removing anything?
Yes. Read the findings and close the tab, or download nothing. The scan itself changes no bytes; only Remove selected data writes a new file.
Why did an attachment survive the sweep?
It was unticked when you ran. Findings you keep stay by design; rescan the download, tick the attachment, and remove again.
Should I scan files I only receive?
Yes, as reconnaissance: the author, software and dates tell you how the file was made before you trust it. Then clean your own files before they travel back.

Last stops before the file travels

Each tool below runs in the browser with no account and checks its own output first. Audit here, destroy page secrets, lock the result, then send.