I built ChromeClaw as a simpler, safer way to use OpenClaw. OpenClaw is powerful, but for many people it’s still more setup than they want — installing software, configuring things manually, and getting comfortable with a more open local environment. I wanted the same core idea of a personal AI assistant, but in a form that’s much easier to start using. ChromeClaw runs entirely inside the browser as a Chrome extension. That makes it easy to install, easy to configure, and cross-platform anywhere Chrome runs. You just install it, add your API key, choose a model, and start chatting. The other big difference is safety. Because ChromeClaw runs inside the browser sandbox, it does not have direct access to your local file system. Your keys and data stay local in the browser, while the agent is constrained by the browser’s security model. That makes it a more approachable option for people who want a personal AI assistant with stronger built-in boundaries. ChromeClaw is also open source under the MIT license, so anyone can inspect it, modify it, and build on top of it. So the problem I was trying to solve was: how do you make OpenClaw-style personal AI accessible to normal users? For me, the answer was: put it in the browser, keep it local-first, remove setup friction, and let the sandbox provide an extra layer of safety.

ChromeClaw

Turn Chrome to your personal Clawbot - private, local-first

Turn Chrome to your personal Clawbot - private, local-first