Browser Operator

Browser Operator turns the browser itself into a place where AI agents do the work. Instead of opening thirty tabs and reading each one, you describe the task and specialized agents collaborate on it, which is what the project means by an intelligent platform for research, analysis and automation.

It is a downloadable application for macOS 10.15 or later and Windows 10 64-bit or later, asking for 8GB of RAM, 16GB recommended, and 2GB of free disk. Configuring a model takes four steps in settings: pick a provider, enter credentials, choose a model, save. OpenRouter signs in through the browser and opens up more than 400 models, OpenAI covers the GPT family, Groq is there for fast inference, and LiteLLM is the route to local models and privacy-sensitive setups.

The result is a browser that handles tasks rather than pages.

• Multi-agent automation: agents work together autonomously on web tasks that span many steps and sites.
• Privacy first: all processing happens locally on your machine, and pairing LiteLLM with Ollama gets you fully offline operation.
• Research and analysis: literature reviews, data collection, competitive intelligence and market research.
• Shopping and price tracking: product comparisons, review analysis and price monitoring over time.
• Business automation: talent sourcing, lead generation and compliance audits.

Releases for both platforms are published on GitHub under the BSD-3-Clause license, with documentation and a Discord community for support. Bug reports, pull requests and doc improvements are all welcome.