What a PDF says about you when you send it
Open the properties of almost any PDF and it will tell you who made it, what software they used, and when. Most people have never looked.
That is fine when the answer is your own name on your own quote. It is less fine in the cases that come up all the time: a document made by saving over an old one and still carrying the previous client’s job title, a tender return that names the person who actually wrote it, a letter that says it was produced on a computer belonging to somebody who left two years ago.
Drop a file in above and you will see exactly what yours is carrying before you decide anything.
How this tool works
The document information and the XMP stream are cleared, and the creation and modification dates are set to a fixed value rather than left to be rewritten with today's, which would put a fresh timestamp on a document somebody is trying to scrub.
One field cannot be cleared and the page says so rather than pretending otherwise: the Producer line, which the underlying library writes on every save whatever it is given. Verified three separate ways. Every PDF this site writes carries it, and a document saying it has been machine-processed is a different order of problem from one carrying an author name and an old job title, which do clear.
It does not touch the pages. Text hidden under a black box, content outside the crop box and anything written into the page itself all survive, because that is a different job and saying otherwise would be a dangerous thing to be wrong about.
Where this bites
- Anything sent to more than one party. Send the same PDF to three contractors and all three can read the title of the job you built it from.
- Anything you did not write yourself. A template off the internet carries the author who published it, so your document goes out with a stranger’s name on it.
- Anything sensitive. Employment paperwork, a grievance, a complaint. The author field is often the one thing nobody thought to check.
- Anything that was scanned. The producer field will happily tell everybody the exact model of the copier in your office.
Clear it, or put your own on
Clearing empties the title, author, subject, keywords and creator, including the ones this tool does not display. Replacing puts your title and name on and empties the rest.
Replacing is the better choice more often than people expect. A document with a proper title shows that title in a browser tab and in most PDF readers, and it is what a document management system will index it under. An empty title means your carefully named file shows up as "Untitled" the moment it leaves your folder.
The one field that will not clear
The producer field comes out naming the library that rewrites the file. That is not a choice, so know that before you rely on this.
Every PDF library stamps its own name on what it saves, and the one behind this tool overwrites that field on every save whatever it is told. Setting it empty, setting it to a space and deleting the entry outright all produce the same result, which is how we know rather than assume. Every other field clears properly, including the two that actually matter: the author name and the title carried over from whatever document this one was built from.
There is a fair argument it should stay. A document that has been through a machine saying so is not really a privacy problem. Somebody's name on a grievance letter is.
What this is not
It is not redaction, and the difference matters enough to say twice.
Metadata is the information about the document. Redaction is about the content in it. Drawing a black rectangle over a name in most editors leaves the text sitting underneath it, selectable and copyable, and clearing the metadata does nothing about that. Cropping a page usually hides the edge rather than removing it. If a document is going out under a legal obligation, or in response to a request, or into a dispute, it needs a proper redaction process and somebody accountable for checking it.
This tool tells you what the properties say and lets you change them. That is all it claims.
Nothing is uploaded, which is the point
A tool that strips private information out of your documents is a strange thing to send your documents to a stranger’s server for.
Every other free metadata remover works by uploading. Yours goes to a machine you do not control, gets processed, and gets deleted at some point they choose, according to a policy you have to take on trust. This one runs in your browser: the file is read, rewritten and saved without a single request leaving your device. You do not have to believe that, either. Open your network tab and watch.
Common questions
Is my PDF uploaded anywhere?
No, and on this tool of all of them that would be an absurd way to do it. The whole job is getting identifying information out of a file, so posting the file to a stranger first would defeat the exercise before it started. Your browser reads it, rewrites it without the metadata and saves the clean copy straight to your machine. Nothing is sent, so there is nothing to trust us about.
Does this redact a document?
No, and that is an important difference. It clears the document properties and the XMP data. It does not remove text hidden under a black box, content cropped off the edge of a page, or anything sitting in the page itself. If you need a document to be safe to release, this is not the tool for it.
What is actually in a PDF that I cannot see?
Usually the author name from whoever’s copy of Word or Acrobat made it, the software and version, the title, and the creation and modification dates. Plenty of PDFs also carry the original document title from a completely different job, because the file was made by saving over an old one.
Does it clear absolutely everything?
Everything except one field. The producer will come out naming the library that rewrote the file, because that library stamps itself on every save and there is no way to stop it. Title, author, subject, keywords and creator all clear properly, and those are the fields that actually carry your name and the old job title. If it matters that nobody can tell a tool touched the document at all, you need a different approach.
Will it change how the document looks?
No. Nothing on any page is touched, so it prints and reads exactly as it did. The file size may shift by a few hundred bytes because the metadata is gone.