Google正在推出Model Context Protocol（MCP）服务器早期测试版本，通过标准化的MCP集成允许第三方AI助手访问和控制Google Home生态系统中的设备12。Claude、OpenClaw、Google Antigravity和Hermes等支持MCP的AI Agent现在可以通过自然语言控制Google Nest门铃、恒温器以及其他Works with Google Home和Matter设备12。这一功能使AI助手能够查看摄像头摘要、监控智能家居活动、控制连接设备，并使用户能够构建自定义仪表板2。
该功能最初面向美国Google Home Premium Advanced订阅用户推出，订阅费为每月20美元2。用户需要创建Google Cloud项目并配置Home MCP，然后向选定的AI Agent提供配置详情2。Google Home应用还推出了电池高效的小部件和其他15项以上的重大功能改进345。
Google has opened its smart home ecosystem to third-party AI agents by integrating the Model Context Protocol (MCP), a standardized framework that allows multiple AI systems to access and control devices within the Google Home environment.12 The new Home MCP integration enables AI agents such as Claude, Antigravity, OpenClaw, and Hermes to interact securely with all devices and event histories in the Google Home ecosystem.126
The capability represents a shift toward what Google calls the "agentic smart home," where users can leverage their preferred AI agents to manage connected devices through natural language commands.234 Users can view camera summaries, monitor smart home activity, control connected devices, and build custom dashboards through their AI agents.2 The feature rollout begins with Google Home Premium Advanced subscribers in the United States, requiring a $20 monthly subscription.2 To enable the functionality, users must create a Google Cloud project, configure Home MCP, and then provide the configuration details to their selected AI agent.2
Alongside the MCP integration, Google Home has received over 15 major feature improvements, including battery-efficient widgets and refined displays for offline devices.345 The new capabilities support all devices within the Google Home ecosystem, including Google Nest doorbells, thermostats, and both "Works with Google Home" and Matter-compatible devices.2
评论
还没有评论，欢迎留下第一条。