A toolbox built on one rule: the file or text you are working on stays on your own device.
Most online utilities work the same way: you paste a JSON payload, a token or a contract, it travels to somebody's server, and something comes back. That is fine for a sample file and a bad idea for a signed agreement, a customer export or a production key — and from the outside you cannot tell which kind of site you are on.
ArrayKit is the other kind. Every tool here runs as code in your browser tab. The document you drop in is read by your own machine, the result is written by your own machine, and neither is sent anywhere. That is also why the tools keep working when your connection drops after the page has loaded.
ArrayKit is built and maintained by Udit Agarwal, a developer who got tired of pasting sensitive data into tools with no explanation of where it went. It is a one-person project, which is worth knowing when you decide how much to trust it — you can see exactly what the pages do, because the work happens in front of you rather than on a server you cannot inspect.
Each tool is a small program compiled into the page. PDFs are parsed, rewritten and rendered locally; images are decoded onto a canvas in your tab; hashes and keys use the cryptography built into your browser. A few tools need extra data the first time you run them — a text-recognition language pack, for instance — and that is the only thing fetched, never your document.
Because everything is local, the practical limits are your device's memory and processor rather than an upload cap or a daily quota. A very large scan simply takes a little longer.
Last updated 2026-09-27.