Google Home Opens Early Access to AI Agent Control

Google Home users will soon be able to run their connected devices through an AI agent. On Tuesday, Google began rolling out early access to a Model Context Protocol (MCP) server for its smart home ecosystem, letting any MCP-compatible agent - among them Claude, Hermes, OpenClaw, ChatGPT, and Google Antigravity - securely interact with smart home devices and pull up their event history.

With the connection in place, users can issue natural language instructions to review camera summaries, keep an eye on home activity, control connected devices, and build custom smart home dashboards.

Setup requires creating a Google Cloud project configured for Home MCP, then handing the MCP configuration details to a chosen agent and asking it to complete the connection. The agent will prompt the user to sign in and grant permissions. A setup guide will also be available in the Google Home Developer Center.

The system covers any device in the Google Home ecosystem, including Google Nest doorbells and thermostats, plus "Works with Google Home" (Matter) devices such as light bulbs.

Google already backs MCP elsewhere in its business, including Google Cloud, its data platforms, developer tools, and Google Workspace. The Google Home support is aimed squarely at consumers experimenting with AI agents for everyday tasks.

Access will reach subscribers of Google Home Premium Advanced in the U.S. starting today and continuing over the coming weeks. That tier costs $20 per month and includes longer event-based video history, descriptive notifications and detailed alerts, video history search tools, and daily summaries. Google declined to comment on whether or when the MCP might reach other subscription tiers or markets. The company says it is gathering feedback from early adopters through its Smart Home for Developers Community.