Google allows AI agents to control Home devices via Model Context Protocol

The update enables users to employ natural language instructions for tasks like reviewing camera summaries, monitoring smart home activity, controlling connected devices, and building custom smart home dashboards. To set up the connection, users must create a Google Cloud project, configure it for Home MCP, and then provide the MCP configuration details to their chosen agent, which will guide them through signing in and granting permissions. A setup guide will be available in the Google Home Developer Center.

The system is compatible with all devices in the Google Home ecosystem, including Google Nest doorbells and thermostats, as well as 'Works with Google Home' (or Matter) devices like light bulbs. Google already utilizes MCP in other areas, including Google Cloud, data platforms, developer tools, and Google Workspace. The integration with Google Home is specifically aimed at consumers experimenting with AI agents for everyday tasks.

Access to MCP is rolling out starting today and continuing over the coming weeks to subscribers of Google Home Premium Advanced in the U.S. This subscription tier costs $20 per month and includes features such as longer event-based video history, descriptive notifications, detailed alerts, video history search tools, and daily summaries. Google has not commented on whether or when MCP will be made available more broadly to other subscription tiers or markets. Google is also collecting feedback from early adopters through its Smart Home for Developers Community.