Document metadata cleaner

Documents quietly record who wrote them, who last edited them, which company's software and licence they came from, and sometimes the full editing history. Journalists, activists and anyone sharing a file can give a lot away without meaning to. This tool shows what is hidden in a PDF or Office file and gives you a cleaned copy.

Nothing is uploaded. Your documents are read by and processed locally in your own browser and never leave your device. This page also tells your browser to block every network request, so it could not send a file anywhere even by mistake.

Your original file is not changed. The tool makes a new copy with "-clean" added to the file name. Once you have opened the copy and checked it, you can delete the original (it still contains the hidden data) and rename the clean copy however you like.

Choose a document (PDF, Word, Excel or PowerPoint), or try a sample.
See what is hidden in it, and what the tool cannot remove.
Download the clean copy and share that one instead.

Options

PDF files: Thorough (rebuilds the PDF, so nothing from older saved versions survives) Quick (edits in place, keeps bookmarks and form fields, may keep older versions)

Also remove comments, sticky notes and markup from PDFs

Replace the names on comments and tracked changes (in Office files) and on comments (in PDFs) with "Author"

Drop documents anywhere in this box

PDF, Word (.docx), Excel (.xlsx) or PowerPoint (.pptx). You can select several at once.

No document handy? See it work with made-up data:

What this removes

Word, Excel and PowerPoint: author, last saved by, company, manager, title and keywords, custom properties, the preview thumbnail, editing time, and the names on comments and tracked changes. Hidden location and camera data inside embedded photos is removed too.
PDF: title, author, subject, keywords, the software that made it, dates, the XMP block (a second copy of all that plus edit history), and hidden data inside embedded photos. Thorough mode also drops attachments, scripts and older saved versions.

What it cannot remove

The content itself. Cleaning does not touch the words, numbers and pictures in the document. If the text names a source or a place, it still does.
Tracked changes and comments. Deleted text in a tracked change is still stored in the file. Accept or reject all changes and delete comments in your word processor first. The tool lists how many it found.
Hidden sheets and speaker notes. A hidden Excel sheet can be unhidden by anyone, and PowerPoint speaker notes travel with the file. The tool tells you when it finds them; delete them yourself.
Black boxes over text do not hide it. In a PDF, a black rectangle drawn over text usually leaves the text underneath, where it can be copied. To redact properly, use the photo redaction tool on a picture of the page, or a proper redaction feature that removes the text.
Thorough PDF mode rebuilds the file, so bookmarks, interactive form fields and accessibility tags are not kept, and links to other parts of the document may stop working. Use quick mode if you need them, and accept that older saved versions may remain.
File names and file dates are outside the document. Rename the clean copy if its name is revealing.
Not supported: older Word, Excel and PowerPoint formats (.doc, .xls, .ppt), OpenDocument files, password-protected PDFs, and scanned pages that contain identifying details in the picture. Some documents carry hidden watermarks or per-copy tracking marks in the text or layout that no tool can find. For high-risk material, share a printed and rescanned copy or ask a trusted helper. See Resources.

Built with JSZip (MIT) and pdf-lib (MIT), both served from this site so nothing is loaded from anyone else.