MiniMax MCP Skill

Overview

Official MiniMax Model Context Protocol (MCP) server for coding-plan users, providing AI-powered search and vision analysis capabilities.

Features

Trigger Scenarios

Use this skill when user says:

• "Search for xxx" / "Look up xxx"
• "Look at this image" / "Analyze this picture"
• "What's in this image" / "Describe this photo"
• "Extract content from URL" / "Fetch this webpage"

Quick Start

1. Get API Key

2. Configure mcporter (Recommended)

# Add MCP server
mcporter config add minimax \
  --command "uvx minimax-coding-plan-mcp -y" \
  --env MINIMAX_API_KEY="your-key" \
  --env MINIMAX_API_HOST="https://api.minimaxi.com"

# Test connection
mcporter list

3. Direct Usage

# Search
mcporter call minimax.web_search query="keywords"

# Analyze image
mcporter call minimax.understand_image prompt="Describe this image" image_source="image-url-or-path"

Usage Examples

See references/examples.md

Environment Variables

Important Notes

⚠️ API Key must match the host region!

If you get "Invalid API key" error, check if your Key and Host are from the same region.

Troubleshooting

• "uvx not found": Install uv - brew install uv or curl -LsSf https://astral.sh/uv/install.sh | sh
• "Invalid API key": Confirm API Key and Host are from the same region
• Image download failed: Ensure image URL is publicly accessible, supports JPEG/PNG/WebP

Related Resources

• GitHub: https://github.com/MiniMax-AI/MiniMax-Coding-Plan-MCP
• MiniMax Platform: https://platform.minimaxi.com (China) / https://www.minimax.io (Global)