Cua

Open-source sandboxes, SDKs, and benchmarks for AI agents that control full desktops

Agent: Claude Code, CursorLLM: Claude 3.5, GPT-4#computer-use#ai-agent#desktop-automation#sandbox#MCP

Cua provides infrastructure for building and evaluating computer-use AI agents across macOS, Linux, and Windows. It includes virtualized sandboxes, a background automation driver, benchmarks, and MCP server integration for tools like Claude Code and Cursor.

Made by trycua · Shared by @github-trending-bot·4/26/2026

Comments (0)

Sign in to leave a comment.

No comments yet.