Navifare MCP Server

Navifare MCP Server

Enables AI assistants to help users find better prices for flights they have already discovered, by comparing prices across multiple booking sources.

Category
访问服务器

README

Navifare MCP Server

Version: 0.1.5

Navifare finds a better price for a specific flight the user already found. Users should provide flight details conversationally, which will be structured into the required format.

Overview

This MCP server wraps the Navifare REST API as MCP tools for flight price comparison. It enables AI assistants like Claude to help users find better prices for flights they've already discovered.

MCP Endpoint

  • Production: https://mcp.navifare.com/mcp
  • Documentation: https://www.navifare.com/mcp (Auto-updates from the MCP endpoint)
  • Learn more about MCP: Model Context Protocol

Tools

format_flight_pricecheck_request

Title: Format Flight Request

Parse and format flight details from natural language text or transcribed image content. Extracts flight information (airlines, flight numbers, dates, airports, prices) and structures it for price comparison. Returns formatted flight data ready for flight_pricecheck, or requests missing information if incomplete.

Annotations:

  • readOnlyHint: true - Tool only formats/parses data, no external calls
  • destructiveHint: false - Tool does not modify or delete data

Input Schema:

  • user_request (required): Flight details in natural language text. Include all available information: flight numbers, airlines, departure/arrival airports and times, dates, prices, passenger counts, and travel class.

Example Request:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "format_flight_pricecheck_request",
    "arguments": {
      "user_request": "I found an Aeroitalia flight XZ2020 from FCO to MXP leaving tomorrow at 19.31 and returning with XZ2021 MXP to FCO next Friday at 10.11. 1 adult, economy, the best fare was 221 EUR"
    }
  }
}

flight_pricecheck

Title: Flight Price Check

Search multiple booking sources to find better prices for a specific flight the user has already found. Compares prices across different booking platforms to find cheaper alternatives for the exact same flight details.

Annotations:

  • readOnlyHint: false - Tool performs searches and may trigger external API calls
  • destructiveHint: false - Tool does not modify or delete data

Input Schema:

  • trip (required): Flight trip details including segments, passengers, and travel class
    • legs (required): Array of flight legs (one for outbound, one for return in round trips)
    • travelClass (required): ECONOMY, PREMIUM_ECONOMY, BUSINESS, or FIRST
    • adults (required): Number of adult passengers (minimum: 1)
    • children (required): Number of child passengers (minimum: 0)
    • infantsInSeat (required): Number of infants requiring a seat (minimum: 0)
    • infantsOnLap (required): Number of infants on lap (minimum: 0)
  • source (required): Source identifier for the query (e.g., "Claude", "MCP", "ChatGPT")
  • price (required): Reference price found by the user (e.g., "84.00", "200.50")
  • currency (required): Three-letter ISO currency code (e.g., "EUR", "USD", "GBP")
  • location (required): Two-letter ISO country code for user location (e.g., "ES", "IT", "US"). If unsure, default to "ZZ"

Example Request:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "flight_pricecheck",
    "arguments": {
      "trip": {
        "legs": [
          {
            "segments": [
              {
                "airline": "XZ",
                "flightNumber": "2020",
                "departureAirport": "MXP",
                "arrivalAirport": "FCO",
                "departureDate": "2025-12-16",
                "departureTime": "07:10",
                "arrivalTime": "08:25",
                "plusDays": 0
              }
            ]
          }
        ],
        "travelClass": "ECONOMY",
        "adults": 1,
        "children": 0,
        "infantsInSeat": 0,
        "infantsOnLap": 0
      },
      "source": "MCP",
      "price": "84",
      "currency": "EUR",
      "location": "IT"
    }
  }
}

Current Limitations

  • Round-trip flights only: One-way trips are not yet supported. Requests with a single leg will return an error immediately.
  • Same origin/destination: Open-jaw trips (returning to a different airport) are not yet supported. The return flight must depart from the outbound destination and arrive at the outbound origin.

Revenue Share

Revenue share available for qualified partners!

Earn commission on flight bookings made through your integration. Revenue share available for qualified partners.

Compliance

This server complies with Anthropic's MCP Directory Policy. See docs/COMPLIANCE.md for details.

Privacy Policy and Terms of Service: https://navifare.com/terms
Contact: contact@navifare.com
Privacy Inquiries: privacy@navifare.com

Environment Variables

The only environment variable you typically need to document for this MCP server is:

  • GEMINI_API_KEY (required): Google Gemini API key for natural language parsing (used when your integration relies on Gemini for text/image understanding).

Examples

See docs/USAGE_EXAMPLES.md for complete usage examples with detailed workflows.

Documentation

Most documentation is in the docs/ directory (and a few files like this README.md and REPOSITORY_STRUCTURE.md live at the repo root):

Getting Started

Reference

Additional Guides

Testing

Test files are located in the test/ directory. See test/README.md for details about available tests.

Running Tests

# Test with MCP Inspector
npx @modelcontextprotocol/inspector node dist/index.js

# Run specific test
node test/test-mcp.js

Scripts

Helper scripts are in the scripts/ directory:

  • scripts/start-local.sh - Start local development server
  • scripts/deploy-to-render.sh - Prepare for Render deployment
  • scripts/start-servers.sh - Start multiple servers
  • scripts/stop-servers.sh - Stop running servers

Repository Structure

navifare-mcp/
├── src/              # TypeScript source code
├── dist/             # Compiled JavaScript (generated)
├── docs/             # Documentation
├── test/             # Test files
├── scripts/          # Helper scripts
└── [config files]    # package.json, Dockerfile, etc.

See REPOSITORY_STRUCTURE.md for detailed structure.

Support

  • General Inquiries: contact@navifare.com
  • Privacy Inquiries: privacy@navifare.com
  • GitHub Issues: https://github.com/navifare/navifare-mcp

Reference

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选