Dragonmcp

What is DragonMCP?

DragonMCP is a Model Context Protocol (MCP) server that bridges AI Agents with local life services in Greater China (Mainland China, Hong Kong) and Asia, enabling tasks like ordering takeout, hailing a ride, checking high-speed rail tickets, and paying utility bills.

How to use DragonMCP?

Install via git clone, npm install, configure a .env file with AMAP_API_KEY, and start the server with npm run dev. Connect to Claude Desktop by adding the built server to claude_desktop_config.json. Docker is also supported.

Key features of DragonMCP

• Real-time MTR train schedule for Island and Tsuen Wan lines
• Amap (Gaode) POI search and walking/driving/transit directions
• Current Hong Kong weather report via HK Observatory
• Mock implementations for Didi, WeChat Pay, Alipay, Meituan, Taobao
• Expansion mocks for Grab, Naver Maps, LINE Pay in Asia

Use cases of DragonMCP

• Ask an AI agent “When is the next train from Admiralty to Central?” and get real-time MTR schedule
• Search for nearby restaurants or POIs using Amap
• Plan walking, driving, or public transit routes with Amap directions
• Get current weather conditions in Hong Kong for trip planning
• Demonstrate mock payment or ride-hailing flows for development

FAQ from DragonMCP

Why do I get "Station not found" for MTR query?

Currently, only Island Line and Tsuen Wan Line stations are supported; check that station names are spelled correctly (e.g., "Admiralty", "Central", "Mong Kok").

How do I get an Amap (Gaode) API Key?

Register at the Amap Open Platform, create a "Web Service" application, and copy the key to your .env file as AMAP_API_KEY.

Can I use DragonMCP for real payments?

No. The current payment tools (WeChat Pay, Alipay) are mocks that return fake data for demonstration only. Do not use real financial credentials.

What are the runtime requirements?

Node.js 18 or higher and npm/yarn. An Amap API key is required for map services.

How do I run DragonMCP with Docker?

Run docker-compose up -d --build to build and start the container; view logs with docker-compose logs -f.