Skip to main content

Documentation Index

Fetch the complete documentation index at: https://platform.minimax.io/docs/llms.txt

Use this file to discover all available pages before exploring further.

We recommend using MiniMax CLI instead of MCP for simpler setup and better experience.

​
Tool Description

Performs image understanding and analysis, supporting multiple image input methods.
ParameterTypeRequiredDescription
promptstring✓Question or analysis request for the image
image_urlstring✓Image source, supports HTTP/HTTPS URL or local file path
Supported formats: JPEG, PNG, GIF, WebP (max 20MB)

​
Prerequisites

1

Get API Key

Visit Token Plan subscription page, subscribe to a plan and get your exclusive API Key.
2

Install uvx

curl -LsSf https://astral.sh/uv/install.sh | sh
For other installation methods, refer to the uv repository.
3

Verify Installation

which uvx
If installed correctly, a path will be shown (e.g., /usr/local/bin/uvx). If you get spawn uvx ENOENT error, you need to configure the absolute path.

​
Use in Claude Code

1

Download Claude Code

Download and install Claude Code from Claude Code official website
2

Configure MCP

Run the following command in terminal, replace api_key with your API Key:
claude mcp add -s user MiniMax --env MINIMAX_API_KEY=api_key --env MINIMAX_API_HOST=https://api.minimax.io -- uvx minimax-coding-plan-mcp -y
3

Verify Configuration

After entering Claude Code, type /mcp. If you can see web_search and understand_image, the configuration is successful.
If you use MCP in an IDE (like TRAE), you also need to configure MCP in the corresponding IDE settings

​
Use in Cursor

1

Download Cursor

Download and install Cursor from Cursor official website
2

Open MCP Configuration

Go to Cursor -> Preferences -> Cursor Settings -> Tools & Integrations -> MCP -> Add Custom MCPCursor MCP Configuration
3

Add Configuration

Add the following configuration to mcp.json file:
{
  "mcpServers": {
    "MiniMax": {
      "command": "uvx",
      "args": ["minimax-coding-plan-mcp"],
      "env": {
        "MINIMAX_API_KEY": "Enter your API key",
        "MINIMAX_MCP_BASE_PATH": "Local output directory path, ensure the path exists and is writable",
        "MINIMAX_API_HOST": "https://api.minimax.io",
        "MINIMAX_API_RESOURCE_MODE": "Optional. Resource delivery method: url or local, default url"
      }
    }
  }
}

​
Use in OpenCode

1

Download OpenCode

Download and install OpenCode from OpenCode official website
2

Configure MCP

Edit the config file ~/.config/opencode/opencode.json, add the following MCP configuration:
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "MiniMax": {
      "type": "local",
      "command": ["uvx", "minimax-coding-plan-mcp", "-y"],
      "environment": {
        "MINIMAX_API_KEY": "MINIMAX_API_KEY",
        "MINIMAX_API_HOST": "https://api.minimax.io"
      },
      "enabled": true
    }
  }
}
3

Verify Configuration

After entering OpenCode, type /mcp. If you can see MiniMax connected, the configuration is successful.