🛫 Fli - Flight Search MCP Server and Library

A powerful Python library that provides programmatic access to Google Flights data with an elegant CLI interface. Search flights, find the best deals, and filter results with ease.

🚀 What makes fli special? Unlike other flight search libraries that rely on web scraping, Fli directly interacts with Google Flights' API through reverse engineering. This means:

• Fast: Direct API access means faster, more reliable results
• Zero Scraping: No HTML parsing, no browser automation, just pure API interaction
• Reliable: Less prone to breaking from UI changes
• Modular: Extensible architecture for easy customization and integration

MCP Server

pipx install flights

# Run the MCP server on STDIO
fli-mcp

# Run the MCP server over HTTP (streamable)
fli-mcp-http  # serves at http://127.0.0.1:8000/mcp/

Connecting to Claude Desktop

{
  "mcpServers": {
    "fli": {
      "command": "/Users/<user>/.local/bin/fli-mcp"
    }
  }
}

Note: Replace <user> with your actual username. You can also find the path to the MCP server by running which fli-mcp in your terminal.

MCP Tools Available

The MCP server provides two main tools:

search_flights Parameters

search_dates Parameters

Quick Start

pip install flights

# Install using pipx (recommended for CLI)
pipx install flights

# Get started with CLI
fli --help

Features

• 🔍 Powerful Search

  • One-way flight searches
  • Multi-city flight searches
  • Flexible departure times
  • Multi-airline support
  • Cabin class selection
  • Stop preferences
  • Custom result sorting

• 💺 Cabin Classes

  • Economy
  • Premium Economy
  • Business
  • First

• 🎯 Smart Sorting

  • Price
  • Duration
  • Departure Time
  • Arrival Time

• 🛡️ Built-in Protection

  • Rate limiting
  • Automatic retries
  • Comprehensive error handling
  • Input validation

CLI Usage

Search for Flights

# Basic flight search
fli flights JFK LHR 2026-10-25

# Advanced search with filters
fli flights JFK LHR 2026-10-25 \
    --time 6-20 \             # Departure time window (6 AM - 8 PM)
    --airlines BA,KL \        # Airlines (British Airways, KLM)
    --class BUSINESS \        # Cabin class
    --stops NON_STOP \        # Non-stop flights only
    --sort DURATION           # Sort by duration

# Alliance + exclude + locale (May-2026 filter additions)
fli flights JFK LHR 2026-10-25 \
    --alliance ONEWORLD \
    --exclude-airlines AA \
    --min-layover 90 \
    --max-layover 360 \
    --currency EUR --language en-GB --country GB

⚠️ Experimental --format json is experimental. The JSON schema may change while the machine-readable CLI contract settles.

# Return machine-readable flight results
fli flights JFK LHR 2026-10-25 --format json

Find Cheapest Dates

# Basic date search
fli dates JFK LHR

# Advanced search with date range
fli dates JFK LHR \
    --from 2026-01-01 \
    --to 2026-02-01 \
    --monday --friday      # Only Mondays and Fridays

⚠️ Experimental --format json is experimental for date searches as well.

# Return machine-readable date search results
fli dates JFK LHR --from 2026-01-01 --to 2026-02-01 --format json

Multi-city Search

# Two-leg multi-city trip
fli multi --leg SEA,HKG,2026-12-26 --leg PEK,SEA,2027-01-02

# Three-leg multi-city trip with filters
fli multi \
    -l SEA,NRT,2026-12-26 \
    -l NRT,HKG,2026-12-30 \
    -l HKG,SEA,2027-01-05 \
    --class BUSINESS \
    --stops 0

CLI Options

Flights Command (fli flights)

Dates Command (fli dates)

Multi Command (fli multi)

MCP Server Integration

Fli includes a Model Context Protocol (MCP) server that allows AI assistants like Claude to search for flights directly. This enables natural language flight search through conversation.

Running the MCP Server

# Run the MCP server on STDIO
fli-mcp

# Or with uv (for development)
uv run fli-mcp

# Or with make (for development)
make mcp

# Run the MCP server over HTTP (streamable)
fli-mcp-http  # serves at http://127.0.0.1:8000/mcp/

Claude Desktop Configuration

To use the flight search capabilities in Claude Desktop, add this configuration to your claude_desktop_config.json:

Location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS)

{
  "mcpServers": {
    "flight-search": {
      "command": "fli-mcp",
      "args": []
    }
  }
}

After adding this configuration:

1. Restart Claude Desktop
2. You can now ask Claude to search for flights naturally:
   • "Find flights from JFK to LAX on December 25th"
   • "What are the cheapest dates to fly from NYC to London in January?"
   • "Search for business class flights from SFO to NRT with no stops"

Python API Usage

Basic Search Example

from datetime import datetime, timedelta
from fli.models import (
    Airport,
    PassengerInfo,
    SeatType,
    MaxStops,
    SortBy,
    FlightSearchFilters,
    FlightSegment
)
from fli.search import SearchFlights

# Create search filters
filters = FlightSearchFilters(
    passenger_info=PassengerInfo(adults=1),
    flight_segments=[
        FlightSegment(
            departure_airport=[[Airport.JFK, 0]],
            arrival_airport=[[Airport.LAX, 0]],
            travel_date=(datetime.now() + timedelta(days=30)).strftime("%Y-%m-%d"),
        )
    ],
    seat_type=SeatType.ECONOMY,
    stops=MaxStops.NON_STOP,
    sort_by=SortBy.CHEAPEST,
)

# Search flights
search = SearchFlights()
flights = search.search(filters)

# Process results
for flight in flights:
    print(f"💰 Price: ${flight.price}")
    print(f"⏱️ Duration: {flight.duration} minutes")
    print(f"✈️ Stops: {flight.stops}")

    for leg in flight.legs:
        print(f"\n🛫 Flight: {leg.airline.value} {leg.flight_number}")
        print(f"📍 From: {leg.departure_airport.value} at {leg.departure_datetime}")
        print(f"📍 To: {leg.arrival_airport.value} at {leg.arrival_datetime}")

Running Examples

Runnable examples live in examples/python/:

# Run examples with uv (recommended)
uv run python examples/python/basic_one_way_search.py
uv run python examples/python/round_trip_search.py
uv run python examples/python/date_range_search.py

# Or install dependencies first, then run directly
pip install pydantic curl_cffi httpx
python examples/python/basic_one_way_search.py

Available Examples:

• basic_one_way_search.py - Simple one-way flight search
• round_trip_search.py - Round-trip flight booking
• date_range_search.py - Find cheapest dates
• multi_city_search.py - Multi-city itinerary across several legs
• advanced_filters_search.py - Alliances, airline exclusions, layovers, locale
• complex_flight_search.py - Advanced filtering and multi-passenger
• time_restrictions_search.py - Time-based filtering
• date_search_with_preferences.py - Weekend filtering
• price_tracking.py - Price monitoring over time
• error_handling_with_retries.py - Robust error handling
• result_processing.py - Data analysis with pandas
• complex_round_trip_validation.py - Advanced round-trip with validation
• advanced_date_search_validation.py - Complex date search with filtering

Examples

Examples are organized by language, with parallel scripts so you can compare the two APIs:

• Python — examples/python/
• TypeScript — examples/typescript/

# Python
uv run python examples/python/complex_flight_search.py

# TypeScript (from examples/typescript, after `bun install`)
bun run multi_city_search.ts

Example Categories:

• Basic Usage: One-way, round-trip, date searches
• Advanced Filtering: Time restrictions, airlines, alliances, seat classes, locale
• Data Analysis: Price tracking, result processing with pandas
• Error Handling: Retry logic, robust error management
• Complex Scenarios: Multi-city, multi-passenger, validation, business rules

Each example is self-contained — change the airports, dates, and filters at the top of the script to fit your search.

TypeScript / JavaScript

Fli is also available as a 1:1 TypeScript port, published to npm as fli-js. Same models, same filter encoding, same direct-API approach.

bun add fli-js   # or: npm install fli-js

import { Airport, FlightSearchFilters, FlightSegment, SearchFlights, SeatType } from "fli-js";

const inDays = (n: number) => new Date(Date.now() + n * 86_400_000).toISOString().slice(0, 10);

const filters = new FlightSearchFilters({
  passenger_info: { adults: 1, children: 0, infants_in_seat: 0, infants_on_lap: 0 },
  flight_segments: [
    new FlightSegment({
      departure_airport: [[[Airport.JFK, 0]]],
      arrival_airport: [[[Airport.LAX, 0]]],
      travel_date: inDays(30),
    }),
  ],
  seat_type: SeatType.ECONOMY,
});

const results = await new SearchFlights().search(filters, { currency: "USD" });

The TypeScript source lives in fli-js/; see the TypeScript Quick Start for the full guide.

Development

# Clone the repository
git clone https://github.com/punitarani/fli.git
cd fli

# Install dependencies with uv
uv sync --all-extras

# Run tests
uv run pytest

# Run linting
uv run ruff check .
uv run ruff format .

# Build documentation
uv run mkdocs serve

# Or use the Makefile for common tasks
make install-all  # Install all dependencies
make test         # Run tests
make lint         # Check code style
make format       # Format code

Docker Development

# Build the devcontainer
docker build -t fli-dev -f .devcontainer/Dockerfile .

# Run CI inside the container
docker run --rm fli-dev make lint test-all

# Or run lint and tests separately
docker run --rm fli-dev make lint
docker run --rm fli-dev make test-all

Running CI Locally with act

To run GitHub Actions locally, install act:

brew install act

# Run CI locally (lint + tests on Python 3.10-3.13)
make ci

# Or run CI inside Docker (no local act installation needed)
make ci-docker

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License — see the LICENSE file for details.