arthurpanhku/DragonMCP

Overview

DragonMCP is a Model Context Protocol (MCP) server that enables AI agents to interact with local life services in Greater China and Asia. It provides standardized interfaces for services such as food delivery, transportation, and other regional utilities. Built in TypeScript, it allows developers to integrate these services into agent workflows using the MCP standard.

Best for

Best for
Developers building AI agents that need to interact with local services in China and Hong Kong

Use cases

• Connect AI agents to local food delivery and ride-hailing APIs in China and Hong Kong
• Enable AI assistants to book services or retrieve real-time local information
• Extend agent capabilities with region-specific service endpoints via MCP

How to use

Install

npm install && cp .env.example .env && npm run dev

Tools exposed

• search_mtr_schedule
• amap_search_poi
• amap_walking_direction
• amap_driving_direction
• amap_transit_direction
• hk_weather_current
• book_taxi_didi
• wechat_pay_create
• alipay_pay_create
• meituan_search_food
• taobao_search_product
• book_ride_grab
• naver_map_search
• line_pay_request

Tested with

Claude Desktop

Example client config

{\n  "mcpServers": {\n    "DragonMCP": {\n      "command": "node",\n      "args": ["/path/to/DragonMCP/dist/server.js"], \n      "env": {\n        "NODE_ENV": "production",\n        "AMAP_API_KEY": "your_amap_api_key_here"\n      }\n    }\n  }\n}

Notes

DragonMCP is a Model Context Protocol (MCP) server that enables AI agents to interact with local life services in Greater China and Asia. It provides standardized interfaces for services such as food delivery, transportation, and other regional utilities. Built in TypeScript, it allows developers to integrate these services into agent workflows using the MCP standard.

23 stars on GitHub. Last updated 2026-03-31. Licensed MIT.

Use cases

• Connect AI agents to local food delivery and ride-hailing APIs in China and Hong Kong
• Enable AI assistants to book services or retrieve real-time local information
• Extend agent capabilities with region-specific service endpoints via MCP

Pros

• Focused on a specific region, offering tailored access to local services
• Uses the standard MCP protocol, making it compatible with many AI agent frameworks
• Open source and written in TypeScript, allowing for easy customization and contribution

Cons

• Limited community adoption with only 23 stars on GitHub
• Only covers Greater China and parts of Asia, not a global solution
• Requires self-hosting and ongoing maintenance to keep service integrations current

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.