OpenBrowse

FAQ

Frequently asked questions about OpenBrowse.

What is OpenBrowse?

OpenBrowse is a free, open source Chrome extension that puts an AI agent in your browser. It can read pages, click, navigate, run code, and connect to external services via MCP — all under your control, with any model you choose.

How is this different from Claude for Chrome, Gemini in Chrome, or Perplexity Comet?

Those are closed, single-vendor products tied to one model. OpenBrowse is open source and model-agnostic — use Claude, GPT, Gemini, a local model via WebLLM, or Chrome's built-in Gemini Nano. You bring your own key, or run with no key at all.

How do I install OpenBrowse before the Web Store listing is live?

Download the latest build from GitHub Releases and load it as an unpacked extension. The full step-by-step is in the Install section of the overview.

When the Chrome Web Store listing is approved, switching over is seamless — both versions share the same extension ID, so your conversations, API keys, MCP connectors, memories, and workspaces all carry over without any manual export/import.

Do I need an API key?

No. OpenBrowse ships with support for Chrome Built-in AI (Gemini Nano) and WebLLM, both of which run fully on-device with no key. If you want a more capable cloud model, you can plug in your own key for OpenAI, Anthropic, Google, or any OpenAI-compatible provider.

Is my data private?

OpenBrowse runs locally as a Chrome extension. Page content and conversations are only sent to the AI provider you pick — if you use a local model, nothing leaves your machine. We don't run a server and don't collect usage data.

Does it work offline?

Yes, when paired with a local model. Chrome Built-in AI and WebLLM both work without an internet connection once the model is downloaded.

Is OpenBrowse open source?

Yes. OpenBrowse is MIT licensed. Contributions are welcome.

On this page