Remove PDF metadata — strip author & XMP in your browser

Clear the author, title, timestamps and creating-application fields from a PDF's Info dictionary and XMP packet — on your machine, with an honest account of what it can and can't reach.

Your file never leaves your device — stripping runs in your browser. 0 bytes uploaded.

Drop a PDF hereTap to choose a PDF

Strips standard document metadata (title, author, producer, dates) and the XMP block, in your browser. Nothing is uploaded.

100% in your browser. 0 bytes uploaded — open DevTools → Network and watch. Your file is read, never sent.

In short

This strips the standard metadata from a PDF — the Info dictionary and the XMP packet — entirely in your browser using pure-JavaScript pdf-lib. That clears author, title, subject, keywords, the creating application and the creation/modification timestamps, then re-saves the file. It's honest about scope: it targets standard document metadata and does not guarantee the removal of every embedded trace. Nothing is uploaded — the Network tab proves it.

Where a PDF hides metadata

A PDF records descriptive data in two standard places. TheInfo dictionary is the classic set of fields:Author, Title, Subject,Keywords, Creator, Producer, and the CreationDate/ModDate timestamps. TheXMP packet is an XML block carrying the same information plus whatever the authoring software chose to add — Acrobat, InDesign, Word export and design tools all write here. This tool clears both, which removes the metadata the overwhelming majority of viewers, search tools and inspectors read.

The honest caveat: what can persist

We won't pretend stripping the standard metadata makes a document forensically anonymous. Emptying the Info dictionary and XMP is the right, expected operation, but a PDF can carry identifying data in places this pass doesn't reach: EXIF data inside embedded images, information tucked into object streams, hiddenlayers or annotations, and — most obviously — anything printed into the visible content of a page itself. If your threat model is real (whistleblowing, legal, safety), the surest route is to flatten the document to images and rebuild it, and to review the result. This tool removes standard metadata cleanly; it does not, and cannot honestly, promise more.

Why no-upload is the whole point here

Removing identifying metadata is a privacy action — so uploading the file to a server to do it would be self-defeating: the original, fully metadata-tagged document would land on infrastructure you can't inspect before anything was stripped. This tool never takes that step. pdf-lib runs in your tab, the file is read from disk and rewritten locally, and no request carries your bytes. Open DevTools, watch the Network tab, and confirm the file stays on your machine.

Encrypted or damaged files

An encrypted PDF must be decrypted before its metadata can be rewritten — the tool detects encryption and points you to Unlock PDF first. If instead a file fails to parse, it may be structurally damaged, and stripping can only rewrite a document it can read. Looks damaged? Diagnose & repair it, then clean the repaired copy here.

FAQ

What metadata does this actually remove?

It clears the two standard places a PDF stores document metadata: the Info dictionary (author, title, subject, keywords, the creating application, and creation/modification timestamps) and the XMP metadata packet (the XML version of the same, plus fields tools like Acrobat and InDesign write). After stripping, those fields are emptied and the file is re-saved. It runs in your browser and is never uploaded.

Does it remove every trace, guaranteed?

No — and we won't claim it does. This targets standard document metadata (Info dictionary and XMP). It does not guarantee removal of every embedded trace: text or EXIF data inside embedded images, information carried within object streams, hidden layers, annotations, or content baked into the page itself can persist. For genuinely sensitive documents, the surest scrub is to export to images and rebuild, or to review the file after stripping. We'd rather tell you the limit than imply a false clean bill of health.

Is my PDF uploaded to clean it?

No — 0 bytes leave your device. Stripping runs in your tab with the pure-JavaScript pdf-lib library; there is no server. Open DevTools, switch to the Network tab, drop your file, and confirm nothing carries it. Removing identifying metadata by uploading the file to someone else's server would defeat the entire point.

Will stripping metadata change how the document looks?

No. The visible pages — text, images, layout — are untouched; only the document's descriptive metadata fields are cleared. The file opens and prints exactly as before, just without the author, timestamps and application fields attached.

Can I clean an encrypted or damaged PDF?

Encrypted files must be decrypted first — the tool detects encryption and points you to unlock. If a file fails to parse, it may be structurally damaged, and stripping can only rewrite a document it can read. Looks damaged? Diagnose & repair it first.

More PDF tools: unlock PDF ·merge PDF ·split PDF ·rotate PDF ·all PDF tools · broken file — repair PDF