ha-mcp-for-xiaozhi

View on GitHub

Homeassistant MCP server for 小智AI Python-based implementation.

158 stars PythonAI & Machine Learning Updated Nov 4, 2025

home-assistantmcpmcp-serverxiaozhi-esp32

Documentation

ha-mcp-for-xiaozhi

Homeassistant MCP server for 小智AI，直连小智AI官方服务器。

插件能力介绍

1.HomeAssistant自身作为mcp server 以websocket协议直接对接虾哥服务器，无需中转

2.在一个实体里同时选择多个API组（HomeAssistant自带控制API、用户自己配置的MCPServer）并将它们一起代理给小智

3.支持同时配置多个实体

功能演示（为爱发电不易，有币投投币、没币点点赞、刷几个弹幕也行）

安装方法：

确保Home Assistant中已安装HACS

1.打开HACS, 搜索 xiaozhi 或 ha-mcp-for-xiaozhi

2.下载插件

3.重启Home Assistant.

配置方法：

[设置 > 设备与服务 > 添加集成] > 搜索“Mcp” >找到MCP Server for Xiaozhi

下一步 > 请填写小智MCP接入点地址、选择需要的MCP > 提交。

注意llm_hass_api 复选框里 Assist 就是ha自带的function，其他选项是你在HomeAssistant里接入的其他mcp server（可以在这里直接代理给小智）

配置完成！！！稍等一分钟后到小智的接入点页面点击刷新，检查状态。

调试说明

1.暴露的工具取决于你公开给Homeassistant语音助手的实体的种类

设置 -> 语音助手 -> 公开

2.尽量使用最新版本的homeassistant，单单看5月版本跟3月版本提供的工具就有明显差异

3.调试时未达到预期，优先看小智的聊天记录，看看小智对这句指令如何处理的，是否有调用homeassistant的工具。目前已知比较大的问题是灯光控制和音乐控制会和内置的屏幕控制、音乐控制逻辑冲突，需要等下个月虾哥服务器支持内置工具选择后可解。

4.如果流程正确的调用了ha内置的function，可以打开本插件的调试日志再去观测实际的执行情况。

Frequently asked questions

What is ha-mcp-for-xiaozhi?

ha-mcp-for-xiaozhi is Homeassistant MCP server for 小智AI Python-based implementation.

How do I install ha-mcp-for-xiaozhi?

Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.

Is ha-mcp-for-xiaozhi open source?

Yes — it is hosted on GitHub at https://github.com/c1pher-cn/ha-mcp-for-xiaozhi and has 158 stars.

Related MCP tools

gpt-researcher

An LLM agent that conducts deep research (local and web) on any given topic and generates a long report with citations. Built for the Model Context Protocol to

24,026 Python

agentaiautomation+8

fastapi_mcp

Expose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth! Python-based implementation. Trusted by 11000+ developers.

11,002 Python

aiauthenticationauthorization+10

ida-pro-mcp

AI-powered reverse engineering assistant that bridges IDA Pro with language models through MCP. Python-based implementation. Trusted by 4100+ developers.

4,123 Python

aiaiagentbinary-analysis+6

web-eval-agent

An MCP server that autonomously evaluates web applications. Python-based implementation. Trusted by 1100+ developers. Trusted by 1100+ developers.

1,191 Python

debuggingdebugging-toolmcp+6

mcp-server-qdrant

An official Qdrant Model Context Protocol (MCP) server implementation Python-based implementation. Trusted by 1000+ developers.

1,043 Python

claudecursorllm+4

mcp-bridge

A middleware to provide an openAI compatible endpoint that can call MCP tools Python-based implementation. Trusted by 800+ developers.

869 Python

aiclaudemcp+5

Run your own MCP server? See who uses it and what to fix.

Measure it with TrackMCP