amap-mcp-demo

# Amap MCP Demo (Node.js + JS) This project demonstrates how to use all available Amap (高德地图) MCP APIs in a Node.js environment, with a simple frontend for interaction. ## Features & APIs Covered - Geocoding (address to coordinates) - Reverse Geocoding (coordinates to address) - IP Location - POI Search (text and around) - POI Detail - Weather Query - Distance Calculation - Route Planning (driving, walking, cycling, transit) - Schema (open Amap app for navigation/taxi) - Personal Map (custom route display) ## Project Structure ``` amap-mcp-demo/ ├── package.json ├── README.md ├── .env.example ├── server.js ├── public/ │ ├── index.html │ └── main.js └── src/ ├── amapClient.js // All MCP API calls └── routes.js // Express routes for each API ``` ## Setup 1. Install dependencies: ```bash npm install ``` 2. Copy `.env.example` to `.env` and fill in your Amap MCP API key. 3. Start the server: ```bash npm start ``` 4. Open [http://localhost:3000](http://localhost:3000) in your browser. ## How to Use - Each API is demonstrated via a simple web interface. - You can test all endpoints directly from the browser or via API calls. --- ## What is Model Context Protocol (MCP)? **Model Context Protocol (MCP)** is an open protocol designed to enable AI models (such as Claude, GPT, etc.) to securely and standardly access external data sources, tools, and services. - MCP is not a proprietary protocol of any single company, but an open standard promoted by AI companies like Anthropic. - The core idea is to let AI models access various "external capabilities" (such as databases, APIs, maps, files, web, etc.) through a standard interface, just like humans can. ### MCP Server and MCP Client - **MCP Server**: A provider of capabilities/data/tools that implements the MCP protocol and can be accessed by AI models via MCP Clients. Examples include: Amap Maps MCP Server, Github MCP Server, Redis MCP Server, Blender MCP Server, etc. - **MCP Client**: The agent for AI models, which can call these servers via the MCP protocol. Examples include: Claude, Cursor, Windsurf, ChatWise, etc. ### Why does AI need MCP? - Traditional AI models can only work with their own knowledge and reasoning, unable to access real-time external information. - With MCP, AI models can "connect to the internet," use tools, query databases, access maps, send emails, and more—greatly expanding their capabilities. - This turns AI into a "super plugin platform" or "digital operator." ### Amap Maps MCP Server and AI - The Amap Maps MCP Server wraps Amap's API capabilities as an MCP protocol service for AI models to call. - This allows AI models (like Claude, GPT, etc.) to use natural language to query routes, POIs, perform geo-analyses, etc., via the MCP protocol. - Thus, AI models gain real-time map capabilities in a standardized, composable way. ### The Significance of mcp.so - [mcp.so](https://mcp.so/) is a marketplace or capability hub for MCP Servers and Clients. - You can find various MCP Servers (Amap Maps, Github, Redis, Blender, Figma, Baidu Map, etc.) and MCP Clients (Cursor, Claude, Windsurf, etc.). - This shows that MCP is becoming the "capability interconnection standard" in the AI ecosystem, enabling seamless collaboration between AI models and external services. ### Summary: The Relationship Between MCP and AI - **MCP is the bridge for AI models to connect with the real world**, enabling AI to not only "chat" but also "do things." - Future AI assistants, agents, and operating systems will use the MCP protocol to call various servers, becoming "super digital employees." - You can think of MCP as the "USB interface for AI"—as long as it supports MCP, AI can plug into all kinds of capability "peripherals." --- References: - [MCP.so Official Website](https://mcp.so/) - [MCP FAQ (English)](https://mcp.so/) (see the bottom of the page) This project is for learning and demonstration purposes. For production use, please secure your API keys and follow best practices.