Architecture map
How the repo is organized
Click systems in the graph to inspect responsibilities, important files, and nearby dependencies.
Architecture map
Use this map to orient around systems, dependencies, and files worth reading first.
Click a system in the map
Explore browser-use/browser-use by selecting services, gateways, clients, and data stores. Each node shows the files, responsibilities, and neighboring systems an agent should keep in scope.
Pan, zoom, and use the minimap to keep large repos navigable.
Sign up to turn selected graph context into repo-aware agent sessions.
Key systems
What an agent should understand first
Command-line interface and task launcher for browser automation
Main agent service that manages the step-loop: state capture, LLM invocation, action execution, and history tracking
Manages CDP connection, tabs, navigation, and WebSocket communication with Chrome
Extracts DOM tree, computes element selectors, and serializes state for LLM consumption
Unified interface to OpenAI, Anthropic, Google, Browser Use Cloud, and other LLM providers
Registers and executes actions: click, type, navigate, scroll, extract, etc.
Low-level browser interaction via CDP: mouse, keyboard, screenshot, element manipulation
Model Context Protocol server exposing Agent as resource for Claude integration
Evidence coverage
What this context is grounded in
Coverage
- Citations
- 8
- Findings
- 5
- Files discovered
- 449
- Architecture systems
- 11
- Architecture relationships
- 14
- Agent graph
- Agent graph entities
- 1,541
- Agent graph relationships
- 3,167