RollingGo-Flight-MCP
RollingGo Flight Search MCP Server. Provides airport search and flight query capabilities for AI Agents and MCP clients via FastMCP.
RollingGo Flight MCP Server
🌐 English | 中文
RollingGo 机票搜索 MCP Server。通过 FastMCP 为 AI Agent 和 MCP 客户端提供机场检索和航班查询能力。
当前版本只暴露 2 个工具:
searchAirports:根据英文城市名、英文机场名或 IATA 代码搜索机场/城市信息。searchFlights:查询指定日期、航线、乘客人数和舱等的航班方案。
暂不实现 checkFlightSeats 和 checkBaggageAllowance。
安装依赖
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt
配置
默认上游 API 地址:
https://mcp.rollinggo.cn
可通过环境变量覆盖:
set ROLLINGGO_API_BASE_URL=https://mcp.rollinggo.cn
HTTP 版本运行时从请求头读取 API Key。调用方需要传入以下任一请求头:
Authorization: Bearer <your_api_key>
或:
X-Secret-Key: <your_api_key>
启动
python server.py
默认监听:
http://127.0.0.1:8000/mcp
MCP 客户端配置示例
{
"mcpServers": {
"RollingGo-Flight-MCP": {
"url": "http://127.0.0.1:8000/mcp",
"type": "streamable_http",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}
searchAirports
keyword 请优先传英文城市名、英文机场名或 IATA 代码,例如 Hangzhou、Chengdu、HGH、CTU。不建议优先传中文城市名。
输入:
{
"keyword": "Hangzhou"
}
上游端点:
POST /api/mcp/airportsearch
searchFlights
输入:
{
"adultNumber": 1,
"childNumber": 0,
"cabinGrade": "ECONOMY",
"fromCity": "HGH",
"toCity": "CTU",
"fromDate": "2026-05-01",
"tripType": "ONE_WAY"
}
上游端点:
POST /api/mcp/flightsearch
字段说明以 rollinggo-readme/FLIGHT-README.md 为准。
Related Servers
tteg
Unsplash without registration. Search and save real stock photos with zero API keys.
Perplexica Search
Perform conversational searches with the Perplexica AI-powered answer engine.
专利大数据服务
Provides comprehensive patent search and statistical analysis for intelligence analysis, technological innovation, and intellectual property management.
Unified Search MCP Server
Provides unified search capabilities across Google Scholar, Google Web Search, and YouTube.
Whois MCP
MCP server that performs whois lookup against domain, IP, ASN and TLD.
Dynamics Partner Advisor
An MCP server for finding and comparing Microsoft Dynamics 365 implementation partners. Features tools to search by industry/region, get detailed partner profiles, and generate personalized shortlists for new projects.
Wikipedia
Retrieves information from Wikipedia to provide context to Large Language Models (LLMs).
VelociRAG
Lightning-fast RAG for AI agents. 4-layer fusion (vector, BM25, graph, metadata), ONNX Runtime, sub-200ms search, no PyTorch.
Releasebot
Releasebot finds and watches release note sources from hundreds of products and companies.
Tavily
Search engine for AI agents (search + extract) powered by Tavily