How it works
Three steps for you. No server anywhere in the loop.
The tool is a static page plus JavaScript. Your PDF is read, parsed, edited and re-saved inside the browser tab you already have open — which is why there is nothing to upload and nothing for us to store.
- No upload endpoint
- No account
- Works offline
The short version
Three steps, zero uploads.
- 01
Drop in your PDF
Drag a file onto the editor or browse for it. The file is read straight into your browser's memory — it is never sent anywhere.
- 02
View and edit the fields
Every metadata field appears in an editable form. Change what you need, clear a field to delete it, or add your own custom properties.
- 03
Download your copy
Save the edited PDF, or strip all metadata in one click. Your original file stays exactly as it was — you always get a new copy.
The longer version
What actually happens when you drop in a file.
Worth reading if “it runs in your browser” sounds like marketing copy. Here is the mechanism, step by step.
- 01
The browser reads the file, not a server
Choosing a file gives the page a File handle from the operating system. We call its arrayBuffer() method to get the raw bytes into a JavaScript variable. That is a purely local operation — there is no network call in it, and no endpoint on this site that accepts a document.
- 02
A PDF library parses those bytes in the tab
The bytes are handed to pdf-lib, a JavaScript PDF toolkit that runs inside the page. It walks the file's object graph, locates the trailer's Info dictionary and the catalog's Metadata stream, and hands back plain values the form can display.
- 03
Your edits are written back into the object graph
Saving sets or deletes each Info key directly, and — for a full removal — drops the XMP stream from the document catalog too. The page content streams are never rewritten, so text, fonts, images and layout come through byte-identical.
- 04
The result becomes a download in local memory
The re-serialised bytes are wrapped in a Blob, given a temporary blob: URL, and clicked as a download link. That URL is revoked immediately afterwards. Nothing is persisted, cached, or transmitted.
Performance
Why it opens instantly.
Reading and writing metadata is a small job: it touches a dictionary near the end of the file and one optional XML stream. It does not require rendering a single page. Building the tool around that fact keeps it fast.
- The page is interactive before any PDF code loads
- The PDF library is imported dynamically the first time you pick a file, so the initial page load stays small. There is no “loading engine…” wait before you can do anything.
- No WebAssembly runtime to download
- Some browser PDF tools ship a multi-megabyte WASM build of a native PDF engine and boot it before showing you a thing. Metadata doesn't need that much machinery.
- Everything is cached after the first visit
- Once the library is in your browser cache, subsequent files are handled with no further downloads at all — which is why the tool keeps working with your network disconnected.
Under the hood
The two places a PDF keeps its metadata.
This is the detail that separates a real cleanup from a cosmetic one. Both stores are handled here.
Store one
Info dictionary
Key/value entries in the file's trailer — what a PDF reader shows under “Document properties”. Editing writes here; removal deletes every entry.
/Title (Q3 Report)
/Author (Jane Doe)
/Producer (Acme PDF 4.1)Store two
XMP packet
An XML stream referenced from the document catalog, often holding a second copy of the same details. Removal deletes the stream reference outright.
<dc:creator>
<rdf:Seq>
<rdf:li>Jane Doe</rdf:li>The viewer shows you the raw XMP when a file has one, so you can confirm for yourself what a “cleaned” document still contains.Read the field reference for what each property means.
Privacy
Your PDF never leaves your device.
This isn't a promise you have to take on faith. There is no server to upload to — the entire tool is static files plus JavaScript that runs in your browser.
- No file uploads — parsing and saving happen in local memory.
- No analytics, no tracking pixels, no advertising cookies.
- No account, no email, no daily quota.
- Works with your network disconnected — try it.
# 1. Open your browser's DevTools → Network tab
# 2. Load a PDF into the editor
# 3. Count the requests that carry your file
→ 0 requests
# Or go fully offline and reload:
# the editor still works, because the work
# was always happening on your machine.See it for yourself.
Open DevTools, load a PDF, and watch the network stay silent.