Google Home now supports the Model Context Protocol (MCP), letting third-party AI agents analyze your smart home data, control connected devices, and build custom dashboards through a standardized integration. The new capability gives tools like Claude, Hermes, and Open Claw direct access to the devices and event history inside a Google Home ecosystem, opening the door to reasoning across what happens in the home rather than just issuing on or off commands.<\/p>\n
Home MCP is an integration that exposes a Google Home account to any AI agent that supports the protocol. Taylor Lehman, group product manager at Google Home & Nest, described the goal in a blog post: the integration “allows any AI agents that support MCP, including Google Antigravity, Claude, Hermes or Open Claw, to securely work with all of the devices and event history in your Google Home ecosystem.”<\/p>\n
In practice, that means an agent can run cross-camera analysis, such as answering what a child did when they got home from school. It can also query device state history to count how many loads of laundry ran last week or measure how long the lights were left on. Agents can interact by voice, sending an audio message through a Google Home speaker when a task finishes, and can create a custom dashboard tailored to the home.<\/p>\n
MCP integration is additive, not a replacement. Google Home smart speakers continue to be controlled by Gemini for Home, which replaced Google Assistant as the interface inside the Home app and on Nest speakers. With MCP enabled, third-party agents gain an additional path to interact with devices through their own interfaces.<\/p>\n
The framing matters: this shifts the smart home from direct commands to systems that respond to context and anticipate needs. A system that actually fixes devices, sharpens automations, and spots usage patterns can do more than one that simply says “I’m sorry, I can’t do that.”<\/p>\n
At launch, access is limited to Google Home Premium Advanced subscribers in the US, priced at $20 per month or $200 per year. Access is rolling out in the coming weeks. Setup requires creating a Google Cloud project and configuring it to use Home MCP.<\/p>\n
Giving an AI agent access to smart locks, thermostats, HVAC systems, and appliances raises immediate questions. Google states that Home MCP enforces rate limits and safety protections, and the system will not allow an agent to open doors. Lehman noted, however, that depending on the agent in use, “connecting it to Home MCP can result in unexpected or even undesired behavior,” and recommended reviewing Google’s developer policies and terms of service.<\/p>\n
Home Assistant, the open-source smart home platform, has already shipped a similar MCP integration. Hands-on testing last year used Claude to vibe-code a smart home setup, with the agent handling troubleshooting, creating automations, designing a dashboard, and configuring advanced setups from natural-language prompts.<\/p>\n
The bigger picture is a developer play. Google launched API access to its smart home in 2024 and more recently turned Gemini for Home into a full-stack AI offering. The MCP announcement adds another entry point: developers can use Google’s Antigravity coding tool to build custom agents that tap into Google Home’s device data and control layer.<\/p>\n
The strategy resembles a B2B2C model: Google supplies the AI and smart home infrastructure while other companies build consumer-facing services and products on top. As long as developers use Google Home, Google wins, regardless of which agent they choose. The comparison in the announcement is AWS for the internet and software businesses.<\/p>\n
Whether developers buy in is an open question. Google has launched and retired multiple smart home platforms over the years, including Android @ Home, Weave, Project Brillo, Works with Nest, and Google Assistant. Developers now have to weigh whether Google has finally settled on a platform it intends to keep.<\/p>\n
The shift is less about an agent toggling lights through a chat interface and more about an agent reading device state and issuing control commands directly. An agent with access to device state and event history can troubleshoot, suggest improvements, and track patterns across devices, capabilities that most smart home apps lack.<\/p>\n
Google Home MCP is a new integration that uses the Model Context Protocol to let third-party AI agents such as Claude, Hermes, and Open Claw analyze a Google Home account’s device data and event history, control connected devices, send voice messages through Google Home speakers, and build custom dashboards.<\/p>\n
No. Gemini for Home remains the assistant for Google Home smart speakers and the Home app, following the retirement of Google Assistant. MCP adds a second path for third-party agents to interact with Google Home devices through their own interfaces.<\/p>\n
At launch, access is limited to Google Home Premium Advanced subscribers in the US, priced at $20 per month or $200 per year, with availability rolling out in the coming weeks. Setup requires creating a Google Cloud project and configuring Home MCP.<\/p>\n