MiniMax MCP Server

Local setup required. This server has to be cloned and prepared on your machine before you register it in Claude Code.
1

Set the server up locally

Run this once to clone and prepare the server before adding it to Claude Code.

Run in terminal
git clone https://github.com/YOUR_USERNAME/mcp-minimax-server.git
cd mcp-minimax-server
npm install
npm run build
2

Register it in Claude Code

After the local setup is done, run this command to point Claude Code at the built server.

Run in terminal
claude mcp add -e "MINIMAX_API_KEY=${MINIMAX_API_KEY}" -e "MINIMAX_GROUP_ID=${MINIMAX_GROUP_ID}" minimax-server -- node "<FULL_PATH_TO_MCP_MINIMAX_SERVER>/dist/index.js"

Replace <FULL_PATH_TO_MCP_MINIMAX_SERVER>/dist/index.js with the actual folder you prepared in step 1.

Required:MINIMAX_API_KEYMINIMAX_GROUP_ID
README.md

Bridge MiniMax AI capabilities to the Model Context Protocol

MiniMax MCP Server

让 AI Agent 通过 Model Context Protocol (MCP) 调用 MiniMax AI 能力

功能概述

mcp-minimax-server 将 MiniMax AI API 桥接到 MCP 协议,使 AI Agent(如 OpenClaw)能够通过标准化的 MCP 工具接口调用 MiniMax 的核心 AI 能力:

工具 说明
minimax_text_understand 图片理解(OCR、图表分析、视觉问答)
minimax_image_generation 文字生成图片
minimax_speech_synthesis 文本转语音

架构设计

AI Agent (OpenClaw)
    │
    │  MCP Protocol (JSON-RPC over stdio)
    ▼
┌─────────────────────────┐
│  mcp-minimax-server     │
│  ┌───────────────────┐   │
│  │  MCP SDK Server   │   │
│  └────────┬──────────┘   │
│           │              │
│  ┌────────▼──────────┐   │
│  │  Tool Handlers    │   │
│  │  ├─ text_understand│  │
│  │  ├─ image_generation│ │
│  │  └─ speech_synthesis│ │
│  └────────┬──────────┘   │
└───────────┼──────────────┘
            │ REST API
            ▼
      MiniMax API

安装

# 克隆或下载
git clone https://github.com/YOUR_USERNAME/mcp-minimax-server.git
cd mcp-minimax-server

# 安装依赖
npm install

# 编译 TypeScript
npm run build

配置

需要设置两个环境变量:

export MINIMAX_API_KEY="your_api_key_here"
export MINIMAX_GROUP_ID="your_group_id_here"

获取方式:

  1. 访问 MiniMax 开放平台
  2. 创建应用获取 API Key 和 Group ID

使用方式

独立运行

npm start

集成到 OpenClaw

在 OpenClaw 配置中添加 MCP 服务器:

{
  "mcpServers": {
    "minimax": {
      "command": "node",
      "args": ["/path/to/mcp-minimax-server/dist/index.js"],
      "env": {
        "MINIMAX_API_KEY": "your_key",
        "MINIMAX_GROUP_ID": "your_group"
      }
    }
  }
}

Claude Desktop 集成

{
  "mcpServers": {
    "minimax": {
      "command": "node",
      "args": ["/path/to/mcp-minimax-server/dist/index.js"],
      "env": {
        "MINIMAX_API_KEY": "your_key",
        "MINIMAX_GROUP_ID": "your_group"
      }
    }
  }
}

工具详解

minimax_text_understand

分析图片内容,支持 OCR、图表解读、视觉问答等。

// MCP 调用示例
{
  name: "minimax_text_understand",
  arguments: {
    image_url: "https://example.com/image.png",
    prompt: "提取图片中的所有文字"
  }
}

minimax_image_generation

根据文字描述生成图片。

// MCP 调用示例
{
  name: "minimax_image_generation",
  arguments: {
    prompt: "一只可爱的橘猫在阳光下打盹,写实风格",
    model: "MiniMax-Image-01"
  }
}

minimax_speech_synthesis

将文本转换为自然语音。

// MCP 调用示例
{
  name: "minimax_speech_synthesis",
  arguments: {
    text: "欢迎使用 MiniMax 语音合成服务",
    voice_id: "male-qn-qingse",
    speed: 1.0,
    pitch: 0
  }
}

支持的声音ID:

  • male-qn-qingse - 男声-青年
  • female-tianmei - 女声-甜美
  • male-qn-ningxuan - 男声-低沉

技术栈

  • 运行时: Node.js 22+
  • 语言: TypeScript (ESNext)
  • MCP SDK: @modelcontextprotocol/sdk ^1.0.0
  • 协议: MCP over stdio (JSON-RPC)

项目结构

mcp-minimax-server/
├── src/
│   └── index.ts          # MCP 服务器主入口
├── dist/                  # 编译输出
├── package.json
├── tsconfig.json
└── README.md

开发

# 调试运行
npm run dev

# 编译
npm run build

# 类型检查
npx tsc --noEmit

License

MIT

Tools (3)

minimax_text_understandAnalyzes image content for OCR, chart interpretation, and visual question answering.
minimax_image_generationGenerates images based on text descriptions.
minimax_speech_synthesisConverts text into natural-sounding speech.

Environment Variables

MINIMAX_API_KEYrequiredAPI key obtained from the MiniMax platform
MINIMAX_GROUP_IDrequiredGroup ID obtained from the MiniMax platform

Configuration

claude_desktop_config.json
{"mcpServers": {"minimax": {"command": "node", "args": ["/path/to/mcp-minimax-server/dist/index.js"], "env": {"MINIMAX_API_KEY": "your_key", "MINIMAX_GROUP_ID": "your_group"}}}}

Try it

Analyze this image at https://example.com/chart.png and summarize the key trends shown.
Generate a realistic image of a cute orange cat napping in the sun.
Convert the following text into a natural-sounding male voice: 'Welcome to the MiniMax speech synthesis service.'
Extract all the text from the provided image URL.

Frequently Asked Questions

What are the key features of MiniMax MCP Server?

Image understanding including OCR and chart analysis. Text-to-image generation capabilities. High-quality speech synthesis with multiple voice options. Standardized JSON-RPC interface via MCP.

What can I use MiniMax MCP Server for?

Automating visual data extraction from charts and documents. Integrating AI-powered image generation directly into chat workflows. Adding text-to-speech functionality to AI agent responses. Building multimodal AI agents that can see, hear, and create.

How do I install MiniMax MCP Server?

Install MiniMax MCP Server by running: git clone https://github.com/YOUR_USERNAME/mcp-minimax-server.git && cd mcp-minimax-server && npm install && npm run build

What MCP clients work with MiniMax MCP Server?

MiniMax MCP Server works with any MCP-compatible client including Claude Desktop, Claude Code, Cursor, and other editors with MCP support.

Turn this server into reusable context

Keep MiniMax MCP Server docs, env vars, and workflow notes in Conare so your agent carries them across sessions.

Need the old visual installer? Open Conare IDE.
Open Conare