How it works

The file stays on disk

A file input gives the page a reference to the file, not a copy on a server. Everything after that is local: the bytes are read into memory, processed, and handed back as a download.

You can watch this yourself. Open the browser’s network tab, run a merge, and you will see no request carrying the file.

Work happens off the main thread

The PDF engine runs in a Web Worker, a separate thread. That keeps the page responsive while a large document is processed, and it is why you can cancel a job: the worker is terminated.

The engine itself is pdf-lib for document structure and pdf.js for rendering pages. Neither is loaded until you actually pick a file, so the page itself stays small.

Why some formats are different

Converting a PDF to Word means reconstructing editable text and layout, which no browser library does well. Those tools run on a server, and the page tells you so before you choose a file.

Encryption is the other special case: it needs a browser feature called cross-origin isolation, so the Protect and Unlock pages run isolated and carry no third-party code.

What we can see

We do not receive your files at all. Aggregate traffic statistics may be collected to know which tools are used; that is described on the privacy page.