Files
wehub-resource-sync 91e75e620b
CI: cua-driver distro-compat matrix / Resolve release version (push) Waiting to run
CI: cua-driver distro-compat matrix / debian:12 (glibc 2.36) (push) Blocked by required conditions
CI: cua-driver distro-compat matrix / fedora:41 (glibc 2.40) (push) Blocked by required conditions
CI: cua-driver distro-compat matrix / rockylinux:9 (glibc 2.34) (push) Blocked by required conditions
CI: cua-driver distro-compat matrix / ubuntu:22.04 (glibc 2.35) (push) Blocked by required conditions
CI: cua-driver distro-compat matrix / ubuntu:24.04 (glibc 2.39) (push) Blocked by required conditions
CI: cua-driver distro-compat matrix / Distro compat summary (push) Blocked by required conditions
CI: Nix Linux Rust source / Nix / compositor build (push) Waiting to run
CI: Nix Linux Rust source / Nix / driver package (push) Waiting to run
CI: Nix Linux Rust source / Nix / Rust unit tests (push) Waiting to run
CI: Rust Linux unit / Rust Linux unit and compile (push) Waiting to run
CI: Rust Windows unit / Rust Windows unit and compile (push) Waiting to run
CI: SPDX Headers / Check SPDX headers (warn-only) (push) Waiting to run
CD: Docs MCP Server / build (linux/amd64) (push) Waiting to run
CD: Docs MCP Server / build (linux/arm64) (push) Waiting to run
CD: Docs MCP Server / merge (push) Blocked by required conditions
chore: import upstream snapshot with attribution
2026-07-13 13:03:19 +08:00

1.6 KiB

cua

The unified Python SDK for Computer-Use Agents.

Installation

pip install cua

Quick Start

from cua import Sandbox, Image, ComputerAgent

# Ephemeral local sandbox with an agent
async with Sandbox.ephemeral(Image.linux(), local=True) as sb:
    await sb.shell.run("uname -a")

    agent = ComputerAgent(model="anthropic/claude-sonnet-4-5", tools=[sb])
    async for response in agent.run("Open the browser and go to example.com"):
        print(response)

What's included

Package Import Purpose
cua-sandbox from cua import Sandbox, Image VM/container sandboxes
cua-agent from cua import ComputerAgent LLM-driven computer-use agent
cua-cli cua command CLI for managing sandboxes and images

cua-agent[cloud] extras are included by default (OpenAI, Anthropic, Gemini API backends).

Extras

pip install cua[omni]       # SOM-based visual grounding
pip install cua[uitars-mlx] # UiTars via MLX (Apple Silicon)
pip install cua[uitars-hf]  # UiTars via HuggingFace
pip install cua[all]        # Everything

Python Version

Requires Python 3.12 or 3.13. For Python 3.11, install cua-sandbox directly.

Telemetry

Cua collects anonymous usage statistics by default. Opt out with:

export CUA_TELEMETRY_ENABLED=false

Or per-instance:

async with Sandbox.ephemeral(Image.linux(), telemetry_enabled=False) as sb:
    ...

agent = ComputerAgent(..., telemetry_enabled=False)