README
Build, benchmark, and deploy agents that use computers
Provides infrastructure to build, benchmark, and deploy AI agents that control computer desktops.
This repo offers an open-source framework for creating agents capable of interacting with desktop environments across macOS, Linux, and Windows. It includes sandboxes for safe execution, SDKs for development, and benchmarks for performance evaluation. Developers can build and test AI agents that automate computer tasks using this unified platform.
Provides infrastructure to build, benchmark, and deploy AI agents that control computer desktops.
Software engineers building AI agents for desktop automation or researchers evaluating AI capabilities in interacting with graphical user interfaces.
Build, benchmark, and deploy agents that use computers