Hopper MCP Server
Enables AI assistants to search flights and hotels, access AI-powered price forecasts, set price alerts, and initiate travel bookings through the Hopper platform. It provides real-time pricing and booking management capabilities directly within the Model Context Protocol.
README
@striderlabs/mcp-hopper
An MCP (Model Context Protocol) server for Hopper — the AI-powered travel booking app with price prediction. This server enables AI assistants to search flights and hotels, get price forecasts, set alerts, and initiate bookings through Hopper's platform.
Features
- Flight search with real-time pricing and AI price predictions
- Hotel search with ratings, amenities, and buy/wait recommendations
- Price forecasting — buy now, wait, or watch recommendations with confidence %
- Price alerts — get notified when prices hit your target
- Flight & hotel booking — full booking flow initiation
- Booking history — view current and past trips
Powered by patchright for stealth browser automation (bypasses bot detection).
Installation
npm install -g @striderlabs/mcp-hopper
Or run directly with npx:
npx @striderlabs/mcp-hopper
Usage with Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"hopper": {
"command": "striderlabs-mcp-hopper"
}
}
}
Config file locations:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Usage with Claude Code
claude mcp add hopper striderlabs-mcp-hopper
Tools
search_flights
Search for available flights with Hopper's price predictions.
origin: "JFK"
destination: "LAX"
departure_date: "2025-06-15"
return_date: "2025-06-22" # optional, omit for one-way
passengers: 2 # default: 1
cabin_class: "economy" # economy | premium_economy | business | first
search_hotels
Search for hotels with ratings, amenities, and price forecasts.
destination: "Paris"
check_in: "2025-06-15"
check_out: "2025-06-22"
guests: 2 # default: 2
rooms: 1 # default: 1
min_price: 100 # optional USD filter
max_price: 300 # optional USD filter
get_price_forecast
Get Hopper's AI buy/wait/watch recommendation with confidence percentage.
trip_type: "flight" # flight | hotel
origin: "NYC" # required for flights
destination: "Tokyo"
travel_date: "2025-08-01"
return_date: "2025-08-14" # optional
set_price_alert
Configure a price drop alert via Hopper app/email.
trip_type: "flight"
origin: "BOS"
destination: "LHR"
travel_date: "2025-07-20"
target_price: 450 # USD
email: "user@example.com"
book_flight
Initiate a flight booking with passenger and payment details.
flight_id: "flight_1" # from search_flights
passenger_first_name: "Jane"
passenger_last_name: "Smith"
passenger_email: "jane@example.com"
passenger_phone: "+1-555-0100"
payment_method: "credit_card"
book_hotel
Initiate a hotel booking. Hopper's Price Drop Guarantee refunds the difference if prices fall.
hotel_id: "hotel_2" # from search_hotels
check_in: "2025-06-15"
check_out: "2025-06-22"
guest_first_name: "Jane"
guest_last_name: "Smith"
guest_email: "jane@example.com"
payment_method: "credit_card"
get_bookings
View current and past bookings from a Hopper account.
email: "jane@example.com"
booking_type: "all" # flight | hotel | all
Example Prompts
- "Search for flights from New York to London in July, one week round trip"
- "What's Hopper's price forecast for hotels in Barcelona next month?"
- "Set a price alert for flights from SFO to Tokyo under $700"
- "Book the cheapest flight you found for Jane Smith, email jane@example.com"
- "Show me my past Hopper bookings for jane@example.com"
Notes
- Booking and booking history features require a Hopper account. The server will guide users to authenticate at hopper.com when needed.
- Price predictions use Hopper's displayed AI recommendations extracted from the live site.
- For best results, use IATA airport codes (e.g. JFK, LHR, CDG) for flight searches.
Development
git clone https://github.com/markswendsen-code/mcp-hopper
cd mcp-hopper
npm install
npm run build
node dist/index.js
License
MIT — Strider Labs hello@striderlabs.ai
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。