cua
Scale computer-use 2.0 with open-source drivers, cross-OS sandboxes, and benchmarks for training, evaluation, and data generation.
- 33k
- Rust
- MIT
- 3 min ago
trycua
Open-source infrastructure for Computer-Use Agents. Sandboxes, SDKs, and benchmarks to train and evaluate AI agents that can control full desktops.
Built from trycua's public GitHub profile, United States of America. Not on Take My Repo yet, so there are no posts or updates here.
Scale computer-use 2.0 with open-source drivers, cross-OS sandboxes, and benchmarks for training, evaluation, and data generation.
A curated list of resources about AI agents for Computer Use, including research papers, projects, frameworks, and tools.
Run Hermes Agent's terminal, file and code tools inside a Cua Space
A lightweight Python library to create, run and manage macOS and Linux virtual machines (VMs) natively on Apple Silicon.
A sample use of Cua Python SDK for running an agent on a cloud sandbox
Generated public projection of the Cua Fleets Terraform provider.
Reusable AI-powered PR end-to-end testing in Cua desktop sandboxes
First-person movement benchmark for cua-driver with a pi-autoresearch loop
Convert PWA manifest URLs to native-looking Android APKs using WebView (no Chrome TWA banners)
Procedural color gradients using Perlin noise mapped through harmonic music interval ratios
Minecraft Java Edition preinstalled in a Windows containerDisk, for cua sandbox how-to testing
Temporary end-to-end test for organization security alert notifications
Electron companion to desktop-test-app — loads arbitrary URLs for cua-sandbox MITM proxy integration tests
Todo-list PWA gym for agent evaluation — installable as Android TWA