Scale computer-use 2.0 with open-source drivers, cross-OS sandboxes, and benchmarks for training, evaluation, and data generation.
trycua/cua is a toolkit for building and testing AI agents that operate computers by seeing screens, clicking controls, and typing across operating systems and virtual machines. It serves agent developers, benchmark authors, and teams generating training data for computer-use models. Catalogue add-ons provide ways to operate Cua drivers and related agent workflows.
Latest release cua-spaces-v0.20.0 — Cua Spaces v0.20.0 · 3 Nov 2026
5 files for Claude Code, Codex, OpenCode and Cursor: poll-github-work, cua AGENTS.md, cua CLAUDE.md, verify-cua-driver and 1 more — 2,632 tokens loaded in every session.
AGENTS.md A 2,621 tok CLAUDE.md A 11 tok .agents/skills/poll-github-work/SKILL.md A 61 tok .cursor/skills/verify-cua-driver/SKILL.md B 60 tok .cursor/skills/verify-cua-spaces/SKILL.md A 97 tok These files are trycua/cua's own configuration — they tell Claude Code, Codex, OpenCode and Cursor how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.