How Utilkits works: your files stay on your device
Utilkits file tools run in your web browser. When you merge a PDF, compress a photo or format JSON, the work happens on your own device, so your file is never uploaded to Utilkits. You can check this yourself in a minute, and every tool page shows a privacy badge that lists the requests the page made while you used it.
What is downloaded
Some tools need an engine or a model the first time you use them: the PDF password tools download a PDF engine, OCR downloads a text-recognition model, and the AI tools download their models from Hugging Face. These are downloads to your browser. Your files are not sent with them, and your browser keeps a copy for next time.
What does contact a server
- The redirect checker sends the address you enter to the Utilkits server, which follows its redirects for you.
- Live-data widgets (Pro) send what you enter to the Utilkits server, which fetches public data for you. See the section below.
- The site uses Google Analytics for page views and tool events: which tool was opened, used or saved from, never file names or contents. When you create an account or start a plan, it also records the sign-in method and the plan. Offline mode on any tool page switches it off for your visit.
- Signing in, plans and saving widgets to an account use the Utilkits account service. The tools themselves do not need an account.
- The Agent API for AI agents is a separate service: files sent to it are processed on Utilkits servers and expire after an hour.
Live-data widgets
Live-data widgets are part of the Pro plan. They fetch public data: exchange rates and precious-metal prices (daily reference rates from Frankfurter), weather (from MET Norway), news headlines (from the feeds you add) and whether a website answers. The Utilkits server makes each request, so the data source sees our server and not your device.
What you enter is sent to the Utilkits server: the currencies, the place as rounded coordinates, the feed addresses or the website address. Your account session goes with it, so the server can check your plan. From there each request carries only the question: no files and no account data are sent to a data source. For the site uptime check and for news feeds, the website you enter receives the request from our server, and its redirects are followed. Each widget page names its source and shows when the data was last updated.
For news headlines, the feed address you type goes to the Utilkits server, which fetches the feed. The publisher sees the Utilkits server, not you. Only the headline, the feed name, the time and the link are shown: no article text and no images.
The privacy badge and offline mode
From the moment you add a file, the badge on each tool page counts the requests the page makes and sorts them into data sent and downloads. Offline mode stops the requests the page itself makes with fetch, XHR and beacons. The page cannot see requests made inside background workers, so the badge says so; the offline test below covers those too.
Check it yourself
- Open a tool page, then open your browser’s developer tools (F12, or Cmd+Option+I on a Mac) and choose the Network tab.
- Add your file and run the tool. Watch the list: requests that download an engine or AI model are fine; look for any request that sends your file out.
- For the strongest test, turn off Wi-Fi after the page has loaded, then run the tool again. If it still works, your file could not have been sent anywhere.