git clone https://github.com/Tongyi-MAI/OpenGCUA.git
# 或浅克隆
git clone --depth 1 https://github.com/Tongyi-MAI/OpenGCUA.git

git clone https://github.com/Tongyi-MAI/OpenGCUA.git
cd OpenGCUA

uv venv --python 3.12
source .venv/bin/activate
uv pip install -e ./packages/sdk -e ./packages/harness/runtime

# 安装原生驱动应用（daemon 已包含在 App 包内）
.venv/bin/mu setup macos
.venv/bin/mu doctor macos

# 安装浏览器支持
.venv/bin/mu setup browser --engine chromium
.venv/bin/mu doctor browser --engine chromium

# 从内置默认配置创建本地私有配置
cp packages/harness/configs/macos_demo.json packages/harness/configs/local.json

{
  "api_key": "YOUR_API_KEY",
  "model_name": "qwen3.8-flash",
  "llm_base_url": "https://dashscope.aliyuncs.com/compatible-mode/v1",
  "llm_protocol": "openai"
}

.venv/bin/mu doctor macos

# 持久可视化终端对话（/exit 或 Ctrl-D 退出）
.venv/bin/python packages/harness/runtime/mac_agent.py

# 强制纯文本输出（适合脚本/日志）
.venv/bin/python packages/harness/runtime/mac_agent.py --ui plain

# 或启动演示工作室 http://127.0.0.1:8800
bash packages/harness/runtime/start_demo.sh

{
  "llm_protocol": "openai-thinking",
  "model_name": "YOUR_GPT_REASONING_MODEL",
  "llm_base_url": "https://api.openai.com/v1",
  "api_key": "sk-..."
}

{
  "llm_protocol": "openai",
  "model_name": "qwen3.8-flash",
  "llm_base_url": "https://dashscope.aliyuncs.com/compatible-mode/v1",
  "api_key": "sk-..."
}

{
  "llm_protocol": "anthropic",
  "model_name": "YOUR_CLAUDE_MODEL",
  "llm_base_url": "https://api.anthropic.com/v1",
  "api_key": "sk-ant-...",
  "anthropic_thinking": {"type": "adaptive"}
}

.venv/bin/python packages/harness/runtime/mac_agent.py

# 把第一个任务作为参数传入
.venv/bin/python packages/harness/runtime/mac_agent.py "帮我预订下周五从上海到北京的机票"

bash packages/harness/runtime/start_demo.sh
# 打开 http://127.0.0.1:8800

import asyncio
from qwen_intelligence import Runtime

async def main():
    async with Runtime() as runtime:
        await runtime.add_macos(name="local-mac")
        async with runtime.session() as mu:
            mac = await mu.register("local-mac", name="mac")
            app = await mac.apps.get("com.apple.TextEdit")
            await mu.emit(await app.ui.ax.observe())   # TextEdit 的无障碍树
            await mu.emit(await app.windows.list())

asyncio.run(main())   # 先打开 TextEdit

<intent>Apply the 29 evidence-based status changes while preserving all original
customer details.</intent>
<code>
todo.update(id="updates", kind="task", content="Applying 29 reviewed status changes ...",
            status="attempted")
r = bash(r"""...
cat > /tmp/apply_lead_statuses.py <<'PY'
...
current=api(p,'/api/crm/workspace')['workspace']
assert current==original,'Workspace changed since original review; stopped before edits.'
for change in changes:
    d=api(p,'/api/crm/leads/'+change['leadId'],'PATCH',{'status':change['to']})
    ...
    assert lead['status']==change['to']
PY
python /tmp/apply_lead_statuses.py""")
if r.ok:
    todo.update(id="status_updates", kind="task",
                content="Applied and API-verified all 29 planned status changes. ...",
                status="claimed_verified")
</code>

OpenGCUA/
├── README.md                    # 主说明（本指南来源）
├── QUICKSTART.md                # 快速开始/配置/排障（492行）
├── ROADMAP.md                   # 路线图
├── CONTRIBUTING.md              # 贡献指南
├── LICENSE                      # Apache-2.0
├── packages/
│   ├── harness/                 # 智能体运行时（macOS 真机）
│   │   ├── runtime/             # 智能体循环与宿主（mac_agent.py 等）
│   │   ├── configs/             # 可运行配置（macos_demo.json）
│   │   ├── examples/osworld/    # OSWorld 集成示例
│   │   └── skills/              # 内置技能（pdf-form、dingtalk-workflow）
│   ├── sdk/                     # Cross-Device SDK
│   │   ├── adapters/macos/      # 原生 macOS 提供者（Rust）
│   │   ├── python/mai_ui/       # SDK 实现
│   │   ├── python/qwen_intelligence/ # 公共别名命名空间
│   │   └── examples/            # 示例程序
│   └── macos/                   # 原生 macOS App（SwiftUI + AppKit）
├── site/                        # 官网源码
│   ├── assets/readme/           # README 官方图
│   └── src/trajectory-assets/   # 轨迹页面资源
└── .github/                     # GitHub Actions 等

@misc{opengcua2026,
  title         = {{OpenGCUA}: Toward a Fully Open Stack for Generalist Computer Use},
  author        = {Hanzhang Zhou and Long Li and Quyu Kong and Lei Wang and
                   Panrong Tong and Long Chen and Tianyu Xia and Gongjie Zhang and
                   Xu Zhang and Chen Liu and Jianan Zhang and Chenglin Cai and
                   Zhengwen Zeng and Steven Hoi and Yue Wang},
  year          = {2026},
  eprint        = {2610.22200},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CV},
  url           = {https://arxiv.org/abs/2610.22200}
}