Sanitize PDF Metadata

Strip author, XMP, photo EXIF, scripts and earlier revisions

Files are processed only in this browser. Nothing is uploaded to a server.

Choose a PDF fileDrop files here, or pick oneOne PDF at a time

Strip author, XMP, photo EXIF, scripts and earlier revisions

Continue in another tool

Items to remove

Items found

The file was analyzed for hidden information. Only the ticked items are removed, and the whole file is rewritten. Text and images on the pages don’t change.

Document info (Info)Not analyzedTitle, author, subject, keywords, creator app, created and modified dates Keep the titleXMP metadataNot analyzedXML stream with the creator app, edit history and document IDsPhoto EXIF · GPSNot analyzedCamera and location data in JPEG photos (pixels stay the same)JavaScript · auto-run actionsNot analyzedActions that run on open, launch files or submit formsAttachmentsNot analyzedFiles embedded in the document and file attachment annotationsComment author infoNot analyzedAuthor names, subjects and timestamps of comments and notesLinks · URL annotationsNot analyzedLinks to external addresses (which may include tracking URLs)

Earlier revisions · deleted objectsNot analyzedAlways removed, because the whole file is rewritten. The document ID also gets a new value.

Your file is ready to download

Continue with this result

Changing a setting hides the previous result. Run it again to save the latest settings.

Remove the password from an encrypted PDF first. Processing focuses on page content, so bookmarks, attachments, forms and signatures may not be preserved. Keep the original and check the result.

About Sanitize PDF Metadata

What leaks from a published PDF is usually not on the page. It is the login name in the Author field, the editing software and history in the XMP packet, GPS coordinates on a phone photo, the names on review comments — and an earlier revision you thought you had fixed.

This tool analyses the file first and shows what it found and how many, then removes the items you tick and writes the whole file out again. Visible text and pictures on the pages are not changed.

What each option actually does

Some items are ticked by default and some are not.

Document info (title, author, creator and producer apps, dates): removed, with an option to keep just the title.
XMP metadata: removed from the document and from pages and images that carry their own.
Photo EXIF and GPS: the APP1 (EXIF/XMP), APP13 (IPTC) and comment segments are cut out of embedded JPEGs. The compressed pixel data after them is untouched, so image quality is identical.
Scripts and auto-run actions, attachments, and author/date fields on annotations: removed.
Links: off by default. When on, link annotations and URL actions are deleted.
Earlier revisions and unreferenced objects: always cleaned. The file is rewritten from scratch with a new document ID, so content left behind by incremental saves, and objects nothing uses any more, are gone.

What it does not remove

Names and numbers visible in the body text, invisible OCR text, values typed into form fields, and text hidden by colour are all outside its scope; use redaction or the hidden-text tool for those. Because the file is rewritten, any certificate-based digital signature becomes invalid, and removing scripts can stop a form's automatic calculations. Encrypted files need their password removed first.

Why a PDF can carry old revisions and orphaned objects at all is explained in the How PDF Files Work guide.

Worked examples

Before putting an event flyer on a website

Input: Analysis: 5 document-info fields (Author “j.smith”), 1 XMP packet, EXIF on 3 photos with GPS on one, an earlier revision present
Output: A new file without author, apps, dates or GPS, with the revisions merged; with “keep title” on, only the title remains

If the flyer was edited and saved once in a way that appended a revision, a deleted sentence can still sit in the old revision. Rewriting the whole file keeps only the objects the current pages use.

How to use it

Choose the PDF and review the analysis (a count for each category).
Tick what to remove; turn on “keep” for values such as the title that should stay.
Run the clean-up, check the new analysis shows nothing left, and save.

Frequently asked questions

Will the file get smaller?

Yes, if it held earlier revisions or many unused objects. If only metadata is removed, the difference is usually a few kilobytes.

Can I use it on a signed contract?

Existing certificate signatures will stop validating. Keep the signed original and clean only the public copy.

Does it remove tracked changes from a Word export?

If they came through as annotations, their author details are removed but the annotations stay. Use the comment removal tool to delete them entirely.

Does it strip metadata from PNG images too?

A PDF does not store PNG files as such. An image added as PNG is kept as Flate-compressed pixels, with no EXIF-style block to begin with. EXIF and GPS survive only when a JPEG was embedded byte-for-byte.

Why does it assign a new document ID?

The PDF document ID links versions of the same document. Leaving the original ID in place lets someone match a public copy to an internal original, so the sanitiser replaces it.

Sources

ISO 32000-1:2008 (PDF 1.7) — Adobe public copy, 7.5.6 Incremental Updates, 14.3 Metadata

Last reviewed: 2026-09-22

Privacy and how data is handled