AR

agentculture/reachy-mini-mcp

开发工具

本页由 AI 翻译,技术细节请以原文为准。 查看原文

34 stars 5 forks 质量 78 趋势 16MIT

一个用于控制Reachy Mini服务器或模拟器的MCP

概览

一个用于使用FastMCP控制Reachy Mini机器人的模型上下文协议(MCP)服务器。 该仓库还包含完整的“对话堆栈”(听觉 + LLM + 会话逻辑),将 Reachy Mini 变成自主对话机器人。专为运行这个完整对话应用而设计。👉 👉 该MCP服务器提供了一套全面的工具,用于控制Reachy Mini的头部运动、天线、摄像头,并执行各种手势和情感表达。 - [ ] 语音:TTS,piper - [ ] 代理层 - [ ] 队列操作支持 - [ ] 中断支持 - [ ] 移动身体支持,reachy-mini python SDK - [ ] 听觉:- [ ] VAD,WebRTCVAD / SileroVAD - [ ] STT,更快的耳语 - [ ] 视觉 - [ ] 内存:- [ ] 文件 - [ ] 向量数据库 - [ ] 向量图数据库 - [ ] 增长,夜间微调支持 - [ ] MQTT 支持 - [ ] 迁移到自主智能仓库 - : 在三维空间(x, y, z)中移动头部,方向(滚转、俯仰、偏航)- : 控制左侧和...

概述(AI 翻译)

一个用于使用FastMCP控制Reachy Mini机器人的模型上下文协议(MCP)服务器。

[!注] 在寻找完整的会话应用吗? 该仓库还包含完整的“对话堆栈”(听觉 + 大型语言模型 + 会话逻辑),将 Reachy Mini 变成自主对话机器人。

Docker 设置专门用于运行这个完整的对话应用。

👉 阅读对话堆栈概览

👉 完整 Docker 安装指南

该MCP服务器提供了一套全面的工具,用于控制Reachy Mini的头部运动、天线、摄像头,并执行各种手势和情感表达。

路线图

MCP

  • [ ] 语音:TTS,吹笛者
  • [ ] 代理层
  • [ ] 队列动作支持
  • [ ] 中断支持
  • [ ] 移动身体支持,reachy-mini python SDK

会话应用

  • [ ] 听力:
  • [ ] VAD,WebRTCVAD / SileroVAD
  • [ ] STT,更快的耳语
  • [ ] 愿景
  • [ ] 记忆:
  • [ ] 文件
  • [ ] 向量数据库
  • [ ] 向量图数据库
  • [ ] 成长,每晚微调支持
  • [ ] MQTT 支持
  • [ ] 转向自主智能仓库

特色

移动与语音控制

本页由 AI 翻译,技术细节请以原文为准。

README

以下为原文 README 节选,完整内容请查看原文。

Reachy Mini MCP Server

A Model Context Protocol (MCP) server for controlling the Reachy Mini robot using FastMCP.

[!NOTE] Looking for the full Conversation App? This repository also contains the full “Conversation Stack” (Hearing + LLM + Conversation Logic) which turns Reachy Mini into an autonomous conversational robot.

The Docker setup is specifically for running this full conversation application.

👉 Read the Conversation Stack Overview

👉 Full Docker Setup Guide

This MCP server provides a comprehensive set of tools to control Reachy Mini’s head movements, antennas, camera, and perform various gestures and emotional expressions.

Roadmap

MCP

  • [ ] Speech: TTS, piper
  • [ ] Agentic layer
  • [ ] Queue actions support
  • [ ] Interruption support
  • [ ] Move body support, reachy-mini python SDK

Conversation app

  • [ ] Hearing:
    • [ ] VAD, WebRTCVAD / SileroVAD
    • [ ] STT, faster-whisper
  • [ ] Vision
  • [ ] Memory:
    • [ ] File
    • [ ] Vector DB
    • [ ] Vector Graph DB
  • [ ] Growth, nightly fine tunes support
  • [ ] MQTT support
  • [ ] Move to autonomous-intelligence repo

Features

Movement & Speech Control

  • Head Control: Move the head in 3D space (x, y, z) with orientation (roll, pitch, yaw)
  • Antenna Control: Control left and right antennas independently
  • Speech (TTS): Make the robot speak while performing actions (via speech parameter)
  • Gestures: Perform predefined gestures (greeting, yes, no, thinking, celebration)
  • Emotions: Express emotions (happy, sad, curious, surprised, confused)
  • Direction Looking: Make the robot look in specific directions (up, down, left, right)

Monitoring & Control

  • State Monitoring: Get full robot state, head state, antenna state, camera state
  • Power Management: Turn robot on/off
  • Emergency Stop: Immediately halt all movements
  • Health Status: Monitor robot health and system status

Advanced Features

  • Command Sequences: Execute multiple robot operations in a single call (NEW!)
  • Single Unified Interface: Access all functionality through one operate_robot tool
  • Repository-Based Tools: Easily extensible tool system with JSON definitions

Prerequisites

在 GitHub 查看完整 README

安装

This server does not publish a one-line install command.

Open the repository installation guide 新手?先看 mcpskill.org 教程 →

配置

{ "mcpServers": { "reachy-mini": { "command": "reachy-mini-mcp", "args": ["serve"], "env": { "REACHY_BASE_URL": "http://localhost:8000" } } } }