GH Repository · trycua
cua
Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.
- stars
- 22,740
- 30-day movement
- +21371/day
- Related entries
- 61
- Connections
- 4
python/uvpythonagent-appnodeHTMLdesktop-automationhacktoberfestwindows-sandboxapplewindowscuamakelumemacosvirtualizationswiftmanusai-agentnode/pnpmcomputer-use-agentvirtualization-frameworkcontainerizationcomputer-useoperator
cua is an open-source computer-use agent stack (trycua/cua) providing drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation. The repository topics point to macOS virtualization via Apple's Virtualization.framework (Lume), Windows sandboxing, containerization, and desktop automation, built with Swift and Python tooling.
You want to run and scale computer-use agents against sandboxed macOS and Windows environments with reproducible evaluation benchmarks.
Use it to
- Sandbox computer-use agents in virtualized macOS/Windows environments
- Run cross-OS agent fleets for data generation
- Evaluate agents with built-in computer-use benchmarks
- Train agents on collected computer-use data
- Automate desktop tasks via the agent stack
For Developers building and benchmarking computer-use AI agents
- Role
- agent-app
- Language
- HTML
- Licence
- MIT
- Forks
- 1,582
- Open issues
- 483
- Last push
- 2026-09-17
- Latest release
- v0.1.0 · 2025-01-31
- Skills shipped
- 1
topicscomputer-useagentsmacoswindowsvirtualizationautomation