SF Tennis Court Booking via Cloudflare MCP Server
Automates tennis court bookings on San Francisco Recreation websites using browser automation, with tools to check availability, initiate booking with SMS verification, and complete bookings.
README
SF Tennis Court Booking via Cloudflare MCP Server
<img width="1079" height="960" alt="image" src="https://github.com/user-attachments/assets/bbec40bb-2ac0-475b-ba31-03a6529fba03" />
Automate tennis court bookings on San Francisco Recreation websites using a Cloudflare MCP (Model Context Protocol) server with browser automation. Never miss your favorite court and court time again!
It uses Stytch to authenticate the server so only certain emails (like mine) can book a court (in my name.)
Demo Video
📹 Watch the demo and a brief project explanation on YouTube
What This Does
This MCP server has 3 tennis booking tools:
- Check Court Availability - See available time slots for any court/date
- Book Court & Request SMS - Automate booking flow up to SMS verification
- Complete Booking with SMS - Finish booking by entering your SMS code
Quick Deploy
This will deploy your tennis booking MCP server to: tennis-booking.<your-account>.workers.dev/sse
Alternatively, clone and deploy locally:
npm create cloudflare@latest -- tennis-booking-mcp --template=cloudflare/ai/demos/remote-mcp-authless
How the tennis booking works
- Check what's available:
check_tennis_courts({
court: "Alice Marble",
date: "2025-07-29",
time: "12:00 PM"
})
- Start booking process (stops at SMS step)
book_and_request_sms({
court: "Alice Marble",
time: "12:00 PM",
date: "2025-07-29"
})
- Manual SMS + Automated Completion (User gets verification code from rec.us, types it in to MCP server message, this tool then runs)
enter_sms_code_and_complete({code: "123456"})
Setup Requirements
You need secrets for your Cloudflare Worker/MCP server:
REC_EMAIL=your-sf-rec-email@example.com
REC_PASSWORD=your-sf-rec-password
Your wrangler.jsonc should have:
"ai": {
"binding": "AI"
},
"browser": {
"binding": "MYBROWSER"
},
Connect to MCP Clients
- Cloudflare LLM Playground
Go to https://playground.ai.cloudflare.com/
Enter your MCP server URL: tennis-booking.<your-account>.workers.dev/sse
Start booking courts with natural language!
- Claude Desktop Install the mcp-remote proxy and update Claude Desktop's config:
{
"mcpServers": {
"tennis-booking": {
"command": "npx",
"args": [
"mcp-remote",
"https://tennis-booking.<your-account>.workers.dev/sse"
]
}
}
}
Now you can chat with Claude: "Book Alice Marble court for tomorrow at 2 PM" and it will handle the entire process!
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。