Yeelight MCP Server

Yeelight MCP Server

Unified MCP-based controller for Yeelight devices enabling natural-language control and status queries.

4
Stars
0
Forks
0
Releases

Overview

Yeelight MCP Server is an intelligent lighting control service built on the Model Context Protocol (MCP). It provides a unified interface for AI assistants, intelligent agents, and other MCP-compatible clients to seamlessly control Yeelight Pro series devices and query their status. The server exposes a Streamable HTTP API with clearly defined interfaces and standardized response formats, enabling easy integration with LLM workflows and agent frameworks such as Claude Desktop, Cursor, and LangChain. It supports core operations including turning devices on/off, adjusting brightness, color and color temperature, curtain control, and multi-channel control, as well as triggering preset scenes for lighting linking and ambiance creation. The design emphasizes privacy and flexibility: you can deploy the open-source code locally on an intranet without cloud dependence, or connect to Yeelight's official cloud service for remote access, discovery, and cross-network operation. Quick start covers two access methods: official remote MCP server or local deployment. Client configuration examples show how to pass Authorization, Client-Id, and House-Id headers. This makes it easy to integrate with popular MCP clients like Cursor, Claude Desktop, and Cherry Studio.

Details

Owner
Yeelight
Language
Python
License
Updated
2025-12-07

Features

Smart Device Control

Control devices with on/off, brightness, color and color temperature adjustments, curtain control, and multi-channel operations.

Status Query Capability

Real-time device status queries for switch state, brightness, color temperature, color value, online status, and other key data.

Scene Mode Execution

Trigger and execute preset scenes for lighting linkage, ambience creation, and coordinated multi-device control.

Easy to Extend and Integrate

Supports local source deployment without cloud dependency and optional connection to Yeelight cloud for remote control and discovery.

AI-Friendly Design

Fully MCP-compliant with streamable HTTP, standardized interfaces, and LLM-friendly invocation; integrates with Claude, Cursor, LangChain.

High Compatibility

Clear interface definitions and standardized responses, naturally compatible with MCP and popular AI agent frameworks.

Audience

AI assistantsInteract with Yeelight devices via natural language using MCP protocol.
Intelligent agentsIntegrate with MCP-enabled agents like Claude Desktop, Cursor, LangChain for device control.
Local deployersDeploy locally for privacy and security, or connect to Yeelight cloud for remote control.

Tags

YeelightMCPMCP ServerSmart LightingPro SeriesAI-friendlyStreamable HTTPLocal DeploymentCloud IntegrationScenesStatus Query