Seats.aero Cloudflare MCP
Enables the deployment of a remote Model Context Protocol server on Cloudflare Workers for flight availability searches via seats.aero. It provides a framework for hosting MCP tools that can be accessed by local clients or the Cloudflare AI Playground.
README
Seats.aero Cloudflare MCP Server
A TypeScript-based, stateless Remote MCP server for interacting with the seats.aero API via natural language. This server runs on the edge via Cloudflare Workers, providing lightning-fast, highly scalable access to flight availability, routes, and bulk searches.
❗ Prerequisites: You will need a seats.aero Pro API key to use this tool.
🚀 Get Started: One-Click Deploy
You can deploy this server directly to your Cloudflare account in one click:
Once deployed, your MCP server will be live at a URL like: https://seats-aero-cloudflare-mcp.<your-account>.workers.dev/mcp
🔐 Crucial Post-Deploy Step: Add Your Secrets
Because this server acts as a secure proxy to the seats.aero API, you must configure your secrets in the Cloudflare Dashboard before the server will function.
- Go to your Cloudflare Dashboard > Workers & Pages >
seats-aero-cloudflare-mcp. - Navigate to Settings > Variables and Secrets.
- Add the following encrypted variables:
SEATS_API_KEY: Your seats.aero Pro API key.MCP_CLIENT_SECRET: A secure password/token you invent to prevent unauthorized access to your edge node.
- Redeploy your worker to inject the new secrets.
🛠️ Tools Available
The server uses strict JSON Schema enums to prevent LLM hallucinations for mileage programs and cabin classes.
get_flights: Get a list of flights. Maps to the cached search endpoint.get_bulk_avail: Retrieve a large amount of availability objects from one specific mileage program. Maps to the bulk availability endpoint.get_routes: Retrieve a list of route objects from one specific mileage program. Maps to the routes endpoint.
💻 Manual Deployment (CLI)
If you prefer to deploy via the command line instead of the one-click button:
- Clone the repository and install dependencies:
git clone [https://github.com/olsonbd/seats-aero-cloudflare-mcp.git](https://github.com/olsonbd/seats-aero-cloudflare-mcp.git)
cd seats-aero-cloudflare-mcp
npm install
- Securely store your sectrets using Wrangler:
npx wrangler secret put SEATS_API_KEY
npx wrangler secret put MCP_CLIENT_SECRET
- Deploy to Cloudflare:
npx wrangler deploy
🤖 Connecting to Claude Desktop
Connect to your remote MCP server from local MCP clients using the mcp-remote proxy.
Open your Claude Desktop configuration file and add the following:
{
"mcpServers": {
"seats-aero": {
"command": "npx",
"args": [
"mcp-remote",
"https://seats-aero-cloudflare-mcp.<your-account>.workers.dev/mcp"
],
"env": {
"MCP_REMOTE_BEARER_TOKEN": "your_mcp_client_secret_here"
}
}
}
}
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。