Find Flights

An MCP server to search for flights.
  • python

168

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "ravinahp-flights-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/Users/YOUR_USERNAME/Code/flights-mcp",
        "run",
        "flights-mcp"
      ],
      "env": {
        "DUFFEL_API_KEY_LIVE": "your_duffel_live_api_key_here"
      }
    }
  }
}

Find Flights MCP Server lets you search and retrieve flight information using Duffel API directly from Claude. This server is ideal for complex itineraries, multi-city trips, and flexible date searches, all powered by conversational prompts and instant flight data.

How to use

You interact with the Find Flights MCP Server through your MCP client to search for flights, view detailed offers, and plan multi-city itineraries. Start by describing your trip in natural language, such as a one-way flight from SFO to NYC on a specific date or a multi-city sequence. The server returns flight options, pricing, connections, and carrier information in structured responses so you can compare quickly. Use it to explore multiple dates or routes without hopping between calendars or complex search forms.

How to install

Prerequisites you need before installing: Python 3.x and an active Duffel API key. You will run the MCP locally and connect it to Claude.

Option A: Install via Smithery (automatic)
```bash
npx -y @smithery/cli install @ravinahp/travel-mcp --client claude

This installs the MCP server for Claude Desktop and wires it into your environment.

Option B: Manual installation (recommended if you want to customize)

git clone https://github.com/ravinahp/flights-mcp
cd flights-mcp

Then install and prepare dependencies using uv (the project uses pyproject.toml for dependency management):

uv sync

Note: We use uv instead of pip in this setup because the project relies on the pyproject.toml configuration.

## Configuration and start

Configure the MCP server in your Claude desktop configuration file to run the local instance. The example below shows the required stdio configuration that launches the local flights-mcp server and passes your Duffel Live API key.

{ "flights-mcp": { "command": "uv", "args": [ "--directory", "/Users/YOUR_USERNAME/Code/flights-mcp", "run", "flights-mcp" ], "env": { "DUFFEL_API_KEY_LIVE": "your_duffel_live_api_key_here" } } }

## Deployment

Build and run the MCP package locally. The process prepares distributions and makes the MCP server available to Claude.

Build steps you will typically perform:
- Sync dependencies and update lockfile
- Build the package
- Start the server via the configured command

## Debugging

For debugging, use the MCP Inspector to monitor requests, responses, and performance in real time. It helps you validate inputs, track errors, and measure latency.

npx @modelcontextprotocol/inspector uv --directory /path/to/find-flights-mcp run flights-mcp


## Available tools

### search\_flights

Search for flights based on parameters such as type (one\_way, round\_trip, multi\_city), origin, destination, departure\_date, and optional filters like return\_date, adults, cabin\_class, departure\_time, arrival\_time, and max\_connections.

### get\_offer\_details

Retrieve comprehensive details for a specific flight offer using its unique ID.

### search\_multi\_city

Search for multi-city itineraries by supplying a list of segments, number of adults, cabin class, and max connections.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational