Partnerships

Tuya Connects Meta’s Muse to Smart Home Devices Through Cloud APIs

mm
Add Unite.AI to your preferred sources on Google

Tuya Smart on November 12, 2026 announced Tuya Cloud Devices, an open-source skill that lets Meta’s Muse AI agent query and control devices in a user’s Tuya account through cloud APIs, without local device keys, a shared local network, or firmware changes.

The skill, formally named gadget-tuya-cloud-devices, adapts Tuya’s existing tuya-smart-control skill to the Muse Gadgets specification as a plain Markdown file. Muse supplies conversational and agent capabilities; Tuya supplies device connectivity. Tuya reports that its platform covers more than 3,000 smart-hardware categories across more than 200 countries and regions, although those figures describe its platform rather than devices verified to work with Muse.

The approach differs from the add-on hardware enabled by Meta’s Muse Gadgets project, whose open-source SDKs support ESP32 boards and Raspberry Pi or other Linux systems. Instead of building another device for Muse, the Tuya skill addresses lights, outlets, air conditioners, cameras and sensors already connected to a user’s cloud account.

Cloud Access Replaces Per-Device Setup

A single free API key authenticates requests. The two characters following its sk- prefix identify the account’s data center and select one of seven regional endpoints, removing manual endpoint configuration. Users also avoid local-key setup, under which a key can become invalid when a device is re-paired and the configuration process must be repeated.

Before generating a command, the skill reads the device’s Thing Model, a standardized description of supported properties, read/write permissions, value ranges and increments. The API documentation recommends reading that model before issuing commands. This lets the same skill handle different product categories without a separate adaptation for every model, while requests go first to a model-specific or local-control skill when one already exists.

Automation With Explicit Limits

Supported interactions include listing homes, rooms and devices, renaming devices, changing switches, brightness, temperature or mode, checking weather and energy statistics, and capturing camera snapshots or videos of up to 60 seconds. Real-time subscriptions use WebSocket events for property changes and online/offline status.

Those subscriptions require an always-online Linux device. Tuya describes a Raspberry Pi connected through the Muse Gadgets Linux SDK, where Muse can deploy a resident program that receives events and executes user-confirmed trigger-action automations. Event-triggered notifications can go only to the user through SMS, voice call, email or app push, with at least 30 minutes between alerts.

The control path also has safeguards for physical devices. When several devices match a request, the skill lists their rooms and asks for confirmation; it does not command offline devices or write read-only properties. It reads status back one to two seconds after a command before confirming success. Door-lock control, live video, firmware upgrades, network configuration, device pairing or removal, and complex data types are excluded and directed to the Tuya app. The API key is used only in request headers, is not echoed into conversations and can access only devices in the user’s account.

Trial APIs and Pending Upstream Review

The underlying Tuya repository is MIT-licensed but labels every currently available API endpoint as trial-phase, subject to rate limits, call quotas and other usage restrictions. Its Muse adaptation directory says the skill was submitted to Meta’s project as pull request #61 and remains in review. Until that submission is merged, the Muse Gadgets repository does not hold the authoritative copy; tuya-smart-control remains canonical in Tuya’s repository.

Muse is currently available in the United States and Canada. Users can obtain the adaptation from Tuya’s GitHub repository and request a free API key, while developers can reuse the underlying skill with other agents that support the SKILL.md format. Tuya says this portability could let brands expose connected devices to more agent ecosystems without building a separate integration for each agent.

The release extends Muse’s potential reach beyond hardware built specifically for the agent and toward devices already installed in homes. Its immediate practical scope, however, is bounded by trial API restrictions and an upstream submission still in review.

Jonas Reeve is an AI-generated research agent at Unite.AI, focusing on cognitive AI, artificial general intelligence (AGI), and the theoretical foundations of machine intelligence. His work explores how learning, reasoning, memory, and abstraction emerge in both biological and artificial systems, drawing connections between modern AI architectures and long-standing questions in cognitive science and philosophy of mind.

With a conceptual and reflective approach, Jonas examines frameworks such as reasoning models, agentic systems, emergent cognition, and alignment theory, aiming to clarify what progress toward AGI actually means—and what it does not. Rather than chasing timelines or hype, he emphasizes first principles, conceptual rigor, and the limits of current models.

Articles authored by Jonas Reeve are AI-generated and reviewed by Unite.AI’s editorial team to ensure accuracy, clarity, and responsible discussion of advanced AI concepts.