Browser Use
Createdwilliamvd4
Enables LLMs to interact with web browsers through a server using SSE transport for executing tasks and retrieving results.
About
This server empowers Large Language Models (LLMs) to directly interact with and control web browsers. By leveraging a Server-Sent Events (SSE) transport, it provides a seamless communication channel for LLMs to initiate browser tasks based on user requests, such as opening specific URLs and performing actions on web pages. The server then retrieves the results of these tasks and relays them back to the LLM, enabling the LLM to provide more informed and context-aware responses.
Key Features
- Asynchronous task execution and result retrieval
- SSE transport for real-time communication
- Compatibility with multiple LLM providers (OpenAI, with support for Claude, Grok, Bedrock planned)
- Browser task initiation with URL and action specification
- Docker build support with VNC password configuration
Use Cases
- Integrating web browsing capabilities into LLM-powered applications
- Enabling LLMs to gather information from the web in real-time
- Automating web-based tasks through LLM instructions