Privacy

The short version: your files never leave your device. Here's exactly what that means, tool by tool.

What runs locally

Every tool in this catalogue processes your files inside your own browser tab — using Web APIs, WebAssembly, and, for a few tools, machine-learning models that run on your device's CPU or GPU. Nothing you drop, record, or type into a tool is sent to a server. Each tool page carries its own Trust Meter, which watches real network activity while you use it and reports the number of bytes uploaded. It should read 0 bytes for the file itself on every tool, for the entire time you use it.

A few tools download a one-time "engine" — a WebAssembly module or a machine-learning model — the first time you use them (for example, the video compressor's ffmpeg engine, or the transcriber's speech model). That download is the tool's own code and model weights, not your data; it's cached by your browser so it only happens once. The Trust Meter doesn't count it as an upload, because nothing of yours is being sent — only something being received.

The one exception: File Drop signaling

File Drop sends files directly between two of your own devices using WebRTC, a peer-to-peer connection built into every browser. To help two devices find each other, a small signaling server (a Cloudflare Worker) relays short connection-setup messages — the technical equivalent of two phones exchanging a phone number before a call. Specifically, it sees:

• A short, randomly generated session code (not tied to any identity)
• WebRTC connection-negotiation messages (SDP offers/answers and ICE candidates) — standard networking metadata, not file content

It never sees the file itself. Each signaling message is capped at 16KB specifically so it's structurally impossible for it to carry file data — the server rejects anything larger. Once the two devices are connected, the file streams directly between them, and the signaling server is no longer involved at all. Nothing is logged or stored after the session ends.

Cookieless analytics

This site uses privacy-focused, cookieless web analytics to understand which pages get visited — no personal data, no cross-site tracking, no cookie banner, because there's no cookie. It reports aggregate counts (which page, roughly when, what country) and cannot identify you individually or follow you across other sites.

No accounts, no uploads, no ads

There's no sign-up, no login, and nothing to lose access to. Nothing you produce with a tool is stored on a server — when you close the tab, it's gone unless you downloaded it. There are no ads and no third-party trackers embedded in any tool page.

Questions

This project is open source — you can read exactly how every tool works, line by line, in its public repository. If something here is unclear, open an issue there.