Question 1

What metadata gets removed?

Accepted Answer

The Info dictionary fields — Title, Author, Subject, Keywords, Producer, Creator, CreationDate, ModDate — are cleared. The XMP metadata stream attached to the document catalog is also deleted when present. After processing, a tool like exiftool or pdfinfo will report empty or zero values for these fields. Custom metadata entries inside the Info dictionary are also cleared. What does NOT get removed: per-object metadata streams attached to individual page objects, fonts, or embedded images — pdf-lib does not walk these. For most use cases (the document properties shown by Preview, Acrobat, and the file inspector) this tool is sufficient.

Question 2

Why would I want to remove PDF metadata?

Accepted Answer

Three common reasons. (1) Privacy — your document was authored on a personal machine and the Author field still shows your full real name; you do not want the recipient to see it. (2) Source protection — a journalist receives a PDF from a source whose name is in the Author field, and the file needs to be shared with editors without exposing the source. (3) Legal review — counsel forwards a contract draft and the Producer / CreationDate would reveal which firm and which billing-eligible day it was drafted on. In each case, the visible content is fine to share; the invisible metadata is the leak.

Question 3

Is the file uploaded anywhere?

Accepted Answer

No. The PDF stays on your device — pdf-lib runs in your browser and writes the cleaned file using only browser memory. Nothing is sent to a server, no temporary copy, no analytics on the file content. This matters: a tool that asks you to upload a PDF to strip metadata is doing the opposite of what you want. The whole point is the document not leaving your control.

Question 4

Does this remove the digital signature or change page content?

Accepted Answer

It does not modify page content or rasterize anything — text, images, layout, and form fields are preserved. Digital signatures will be invalidated because rewriting the document inevitably changes the byte sequence the signature was computed over. If signature integrity matters for your workflow, strip metadata before signing, not after. The same applies to certifying signatures and timestamps.

Question 5

What about the XMP metadata stream specifically?

Accepted Answer

Modern PDFs (those produced by Acrobat, InDesign, Word, LaTeX since the mid-2000s) often carry an XMP metadata stream in addition to the legacy Info dictionary. The XMP block can include camera info, GPS coordinates from a phone scan, application history, even thumbnail data. This tool removes the XMP stream object from the document catalog when present. On unusual PDFs where pdf-lib cannot delete the XMP stream cleanly, the tool will report the limitation rather than silently leave it in place.

Question 6

Will the file size change?

Accepted Answer

Slightly smaller in most cases — the Info dictionary entries and the XMP stream are gone. The page content streams (which carry the text and images) are not recompressed, so the saving is only the metadata bytes. For a typical office PDF, expect a few kilobytes of reduction; for a large scanned PDF, the percentage change is negligible. If you also want to shrink the file, run /compress after removing metadata.

Question 7

Can I check what metadata was in the file before / after?

Accepted Answer

Yes. The tool shows the metadata it detected before processing — Title, Author, Producer, Creator, Subject, Keywords, CreationDate, ModDate, and whether an XMP stream was present. After saving, all visible fields are blanked. To verify externally, run `exiftool yourfile.pdf` or `pdfinfo yourfile.pdf` — both should report the cleared values. If your auditor or recipient runs their own check, the result will match.

Question 8

What about anti-forensic redaction of page text?

Accepted Answer

This tool is for metadata only — the document properties surfaced by every PDF reader. It does not redact visible text on the page. If you want to permanently cover text inside a page, use /redact-pdf for visual coverage, or a desktop tool like qpdf/ghostscript pipelines or Acrobat Pro Redact for content-stream removal. For most leak scenarios both layers matter — strip metadata here, then redact visible content separately.

How it works

Who this is for

Related Tools