MiniMax-Coding-Plan
Specialized MiniMax Model Context Protocol (MCP) server designed for coding-plan users, featuring AI-powered search and vision analysis APIs optimized for code development workflows
claude mcp add --transport stdio minimax-ai-minimax-coding-plan-mcp uvx minimax-coding-plan-mcp -y \ --env MINIMAX_API_KEY="insert-your-api-key-here" \ --env MINIMAX_API_HOST="https://api.minimax.io"
How to use
The MiniMax Coding-Plan MCP server provides AI-powered search and vision-analysis capabilities tailored for coding workflows. It exposes tools like web_search for rapid discovery of code examples, references, and documentation, and understand_image for analyzing screenshots or diagrams related to coding tasks. When connected through an MCP client (such as Claude Desktop, Cursor, Windsurf, or OpenAI Agents), you can issue requests to the MiniMax MCP server and receive structured responses that help you locate relevant code snippets, analyze UI flows, or extract insights from images of code or UI mockups. The server is designed to be run locally or in the cloud and supports two transport modes: stdio (local execution) and SSE (network-based deployment). Ensure your MCP client is configured with the correct API host and API key for your region to avoid authentication errors.
To use the server from a client, configure the MCP server entry in your client with the MiniMaxCodingPlan server name, the uvx command, and your API credentials. After starting the server, invoke the available MCP tools (e.g., web_search and understand_image) through your client’s MCP interface. The tools integrate with your coding environment to accelerate searches, summarize results, and reason about visuals in code development tasks.
How to install
Prerequisites:
- Python with uv (or a compatible Python environment) and curl installed on your system
- Access to a MiniMax API key and a region-appropriate API host
Step-by-step installation:
- Install the uv package manager (Python-based) for your platform:
- macOS/Linux: curl -LsSf https://astral.sh/uv/install.sh | sh
- Windows: follow the instructions on the uv repository (https://github.com/astral-sh/uv)
- Ensure Python is available in your PATH and verify uv is installed:
- uv --version
- Create and configure your MCP server entry (example below) with your API credentials. You can copy this into Claude Desktop or your preferred MCP client configuration file.
- Start the MCP server via your client or directly using the uvx command once configured.
Notes:
- Use a region-matched MINIMAX_API_HOST and MINIMAX_API_KEY to prevent authentication errors.
- The example configuration uses the MiniMaxCodingPlan server name and the minimax-coding-plan-mcp MCP package/task.
Additional notes
Tips and common issues:
- Ensure the API host matches the region where your API key is issued. Mismatched hosts can cause Invalid API key errors.
- If you switch regions, update MINIMAX_API_HOST accordingly.
- For Claude Desktop users, ensure Developer Mode is enabled if required by your environment.
- The server supports two transports: stdio (local) and SSE (network). When deployed in the cloud, prefer SSE for remote access.
- The MCP server is designed to integrate with coding-focused tools like web_search and understand_image to streamline coding tasks; combine results with your existing IDE or editor for best productivity.
Related MCP Servers
mcp
Official MCP Servers for AWS
cursor-notebook
Model Context Protocol (MCP) server designed to allow AI agents within Cursor to interact with Jupyter Notebook (.ipynb) files
PixVerse
Official PixVerse Model Context Protocol (MCP) server that enables interaction with powerful AI video generation APIs.
fastmcp-threatintel
AI-Powered Threat Intelligence MCP tool
minimax_search
MiniMax Search is an MCP (Model Context Protocol) server that provides web search and browsing capabilities.
MCP-Mathematics
A comprehensive Model Context Protocol (MCP) server that turns any AI assistant into a powerful mathematical computation engine.