ottercraft.devGuide · Privacy
How to tell whether an online tool uploads your files
"Your files are safe with us" is printed on every PDF-merging website. Some of them mean "we process it in your browser and never see it". Others mean "we upload it, keep it for a while, and promise to delete it". You can check which one you're looking at in about two minutes, without trusting anyone's wording.
Why it matters
The documents people run through these tools are exactly the ones they'd rather not hand out: contracts, payslips, scans of ID, medical letters, bank statements. Once a file has been uploaded, it exists on a machine you don't control, under a retention policy you can't verify, in a jurisdiction you may not know. Even with the best intentions on the company's side, that's a risk you don't need to take for a task your browser can do on its own.
The check: watch the network tab
- Open the tool in Chrome, Edge, Firefox or Safari on a computer.
- Open developer tools. Press F12, or Ctrl+Shift+I on Windows and Cmd+Option+I on a Mac. In Safari, enable the Develop menu in Settings first.
- Click the Network tab. Tick "Preserve log" if you see it.
- Now use the tool: add a PDF, merge, download.
- Look at the list of requests. Sort by size, or look at the Type column.
If you see a request with a size roughly the same as your file, usually a POST or PUT, with a name like upload, process or a long random id, your file was uploaded. If the only requests are the page's own code (.js, .css, .wasm, fonts, a favicon) and nothing large appears when you add a file, the work is happening in your browser.
Other tells
- A progress bar while "uploading" before you can do anything. Local tools don't upload, so there's nothing to wait for besides drawing thumbnails.
- "Files are deleted after 1 hour / 24 hours." That sentence only makes sense if the files were stored somewhere. A local tool has nothing to delete.
- A file-size limit like "up to 50 MB on the free plan" points at server storage and bandwidth being paid for. Local tools are limited only by your device's memory.
- It works offline. Turn off Wi-Fi after the page has loaded and try to use it. A local tool keeps working; a server-based one stops.
- The download URL. If the finished file downloads from an address like
https://cdn.example.com/files/8f2a…pdf, it was on their server. If it downloads from ablob:address, it was made in your browser.
What "processed in your browser" actually means
Modern browsers can run substantial programs. A PDF library compiled to JavaScript or WebAssembly can open, render and rewrite a document entirely on your machine; a WebRTC connection can move a file between two devices without a server in between. When a site is built this way, the server's only job is to hand you the code once. After that, your file is read by that code, on your device, and the result is written back to your device. There's no upload step to secure because there's no upload.
That's how the tools on this site work: OtterPDF for merging, splitting and rotating PDFs, and OtterDrop for moving files between devices. Run the network-tab check on them too. You should see the page load its own code and nothing else, no matter how many files you add.