Loading article…
Google Home now supports third-party AI agents like Claude and OpenClaw for smart home control. The feature is rolling out to US Premium Advanced users
Google Home has begun rolling out early access for third-party AI agent support, allowing services like Anthropic's Claude and the open-source OpenClaw to control smart home devices and access event history [1, 2]. This move expands Google Home's capabilities beyond its own Gemini AI, enabling more complex natural language commands and custom dashboards for users [2].
| At a glance | |
|---|---|
| Product | Google Home |
| New Feature | Third-party AI agent support |
| Initial Availability | US Google Home Premium Advanced subscribers |
| Subscription Cost | $20 per month [1] |
The new functionality leverages Google's Model Context Protocol (MCP) server, which has previously been used in Google Cloud, data platforms, developer tools, and Google Workspace [1]. With MCP, AI agents that support the protocol can securely interact with any device in the Google Home ecosystem, including Google Nest products and "Works with Google Home" (Matter) devices [1]. This allows users to review camera summaries, monitor smart home activity, and control connected devices using natural language instructions from their chosen AI agent [1].
The rollout began on Wednesday for Google Home Premium Advanced subscribers in the U.S., a $20-per-month tier that includes features like extended video history and detailed alerts [1, 2]. Google has not commented on whether or when MCP support will be extended to other subscription tiers or markets [1]. The company is soliciting feedback from early adopters through its Smart Home for Developers Community [1].
Google Home is the first of the major smart home platforms—Apple Home, Amazon Alexa, and Google Home—to offer this level of third-party AI integration [2]. Smaller, open-source platforms like Home Assistant have offered similar capabilities for years, allowing users to integrate advanced AIs like OpenClaw and Claude for localized control [2]. OpenClaw, for instance, is an open-source AI agent designed to run on users' machines, managing tasks like email and calendar [3].
The integration allows for more advanced automations and personalized control than Google's native Gemini for Home might provide [2]. For example, users could link thermostat actions to specific local weather reports or create custom dashboards [2]. However, the expansion of access to third-party AI agents also introduces potential security vulnerabilities, such as prompt injection attacks, which Google has previously encountered and patched in early versions of Gemini [2].
This move by Google opens its smart home platform to a wider array of AI capabilities, potentially enhancing user control and customization, but also introduces new considerations for security and platform stability.
Coverage is mostly measured — 297 of 300 reports stay neutral.
Every Monday — the token unlocks, Fed dates & catalysts set to move crypto and markets this week. So you’re never blindsided.
Free · 3-min read · one-click unsubscribe
AI-assisted synthesis by the TrendWatcher Editorial Desk · sourced from 3 outlets · Sep 17, 2026 · How we report
As of late 2026, Google Home supports Anthropic’s Claude, the open-source agents OpenClaw and Hermes, and the Google Antigravity coding platform.
Google shortened the Chrome update cycle from four weeks to two weeks to minimize the window of time between vulnerability disclosure and patching, helping to defend against fast-moving, AI-enabled cyber threats.
Yes, as of late 2026, users must have a $20 Google Home Premium Advanced subscription to access third-party AI agent capabilities.
Integrating third-party AI into Google Home introduces potential vulnerabilities such as prompt injection, where malicious instructions could be hidden in emails or calendar events to control home devices without user knowledge.