Flighty MCP Server

Flighty MCP Server

Connects AI assistants to the Flighty flight tracking app for querying flight status, delay forecasts, stats, and adding/removing flights via natural language.

Category
Visit Server

README

Flighty MCP Server

A zero-config MCP server that connects AI assistants to the Flighty flight tracking app. Ask about your flights, check statuses, look up delay forecasts, and add or remove flights — all through natural conversation.

Just install Flighty on your Mac and connect the server. It reads all credentials and configuration directly from the app — no API keys, no tokens, no setup.

What it does

This server gives any MCP-compatible AI assistant (Claude, etc.) full read/write access to your Flighty data:

  • "What's my current flight?" — shows in-progress, recently landed, and about-to-depart flights
  • "What's my next flight?" — returns the soonest upcoming flight first
  • "How often is AA179 delayed?" — historical on-time performance with percentage breakdowns
  • "Show my flight stats for 2025" — total flights, miles, top airlines, top routes
  • "Add DL10 on April 20" — adds the flight to your Flighty account, syncs to all devices within seconds
  • "Remove that flight" — deletes it from your account across all devices
  • "Are any friends flying soon?" — checks connected friends' upcoming flights
  • "What version of the Flighty tool is this?" — server info, capabilities, and version

Read operations query Flighty's local SQLite database directly (fast, offline-capable). Write operations call Flighty's API so changes sync to your phone, watch, and widgets. All responses include contextual metadata (sort order, timestamp format) to help AI assistants interpret results correctly.

Requirements

  • macOS — Flighty stores its database in the macOS app sandbox
  • Flighty macOS app — installed and signed in
  • Flighty Pro — required for the add/remove flight features
  • Node.js 18+ (only for manual install; the .mcpb bundle includes its own runtime)

Installation

Claude Desktop (.mcpb) — recommended

Download flighty-mcp-server.mcpb from the dist/ folder and double-click it. Claude Desktop installs it automatically. No configuration needed.

Claude Code

git clone https://github.com/CPLX/flighty-mcp-server.git
cd flighty-mcp-server
npm install
npm run build

Add to your MCP config (~/.claude/settings.local.json or project .mcp.json):

{
  "mcpServers": {
    "flighty": {
      "command": "node",
      "args": ["/absolute/path/to/flighty-mcp-server/build/index.js"]
    }
  }
}

Claude Desktop (manual)

Build from source as above, then add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "flighty": {
      "command": "node",
      "args": ["/absolute/path/to/flighty-mcp-server/build/index.js"]
    }
  }
}

Restart the client after adding the config.

Tools

14 tools organized into four categories.

Flight Management

Tool Description
flighty_list_flights List your flights — upcoming (sorted soonest first), by year, or all (sorted most recent first)
flighty_current_flights Active travel context: flights departing within ±24 hours of now
flighty_get_flight Get details for a single flight by UUID or flight number (e.g., "AA179")
flighty_search_flights Search your flight history by airline, airports, or date range
flighty_add_flight Add a flight by code and date (e.g., "DL10" on "2026-04-20"). Syncs to all devices
flighty_remove_flight Remove a flight by UUID. Permanent deletion across all devices

Flight Intelligence

Tool Description
flighty_get_flight_status Current status (scheduled/delayed/in_air/landed/cancelled), gate, delay, weather
flighty_get_delay_forecast Historical on-time stats: % early, on-time, late, cancelled, diverted
flighty_get_flight_stats Aggregate stats: total flights, miles, circumnavigations, top airlines and routes
flighty_get_connections Layover info for connecting flights: airports, duration, minimum connection time

Social and Reference

Tool Description
flighty_list_friend_flights Connected friends' flights, filterable by name, upcoming, or year
flighty_search_airports Search airports by IATA/ICAO code, city, or name
flighty_search_airlines Search airlines by IATA/ICAO code, name, or alliance

Server Info

Tool Description
flighty_about Version, author, repository, capabilities, and requirements

Tool details

flighty_list_flights

Lists your flights with smart sort order based on the filter:

  • upcoming_only=true: sorted soonest first — the first result is your next flight
  • No filter or year: sorted most recent departure first (natural browsing order)

Parameters:

  • upcoming_only (boolean, default: false) — only flights with future departures, sorted soonest first
  • year (integer, optional) — filter to a specific year
  • limit (integer, 1-200, default: 50)
  • offset (integer, default: 0)

For flights currently in progress or recently landed, use flighty_current_flights instead.

Includes both auto-detected commercial flights and manually-entered flights (private/charter operators not in Flighty's commercial database, or commercial flights the user added by hand). Same applies to flighty_search_flights, flighty_get_flight, flighty_current_flights, flighty_get_flight_status, flighty_get_delay_forecast, flighty_get_flight_stats, and flighty_get_connections.


flighty_current_flights

Returns flights departing within 24 hours in either direction from now. Captures in-progress flights, recently landed flights (baggage claim, layovers), and flights about to depart.

No parameters. Sorted by departure time ascending (earliest first).


flighty_get_flight

Looks up a single flight by UUID or flight number.

Parameters (provide one):

  • flight_id (string) — internal Flighty UUID
  • flight_number (string) — e.g., "AA179", "DL10". If flown multiple times, returns the most recent instance

flighty_search_flights

Searches your flight history with combinable AND filters. Sorted most recent departure first.

Parameters (all optional):

  • airline — IATA code ("AA") or partial name ("American")
  • departure_airport — IATA code ("JFK") or city ("New York")
  • arrival_airport — IATA code ("SFO") or city ("San Francisco")
  • after — flights departing on or after this date ("2025-01-01")
  • before — flights departing on or before this date ("2025-12-31")
  • limit (integer, 1-200, default: 50)

flighty_get_flight_status

Returns the operational status of a flight that is already in the user's Flighty database. This only works for flights the user has added to Flighty — it cannot look up arbitrary flight numbers.

Parameters:

  • flight_number (string) — a flight in the user's database, e.g., "UA194"

Returns: Status ("scheduled", "delayed", "in_air", "landed", "cancelled"), departure/arrival delay in minutes, gate assignments, baggage belt, weather conditions, aircraft type, and tail number.

Data freshness depends on the Flighty app's last sync — this does not make live API calls for status.


flighty_get_delay_forecast

Historical on-time performance for a flight that is already in the user's Flighty database. Flighty attaches delay forecast data when a flight is added to the user's account — this tool reads that stored data. It cannot look up forecasts for arbitrary flight numbers that aren't in the database.

Parameters:

  • flight_number (string) — a flight in the user's database, e.g., "AA179"

Returns: Number of observations (sample size), mean delay in minutes, and percentage breakdowns: early, on-time, late (15/30/45+ min), cancelled, diverted. Returns null if no forecast data is available.


flighty_search_airports

Searches Flighty's airport database. Exact IATA/ICAO matches are prioritized over fuzzy name/city matches.

Parameters:

  • query (string) — IATA code, ICAO code, city name, or airport name
  • limit (integer, 1-50, default: 10)

flighty_search_airlines

Searches Flighty's airline database.

Parameters:

  • query (string) — IATA code, ICAO code, airline name, or alliance name
  • limit (integer, 1-50, default: 10)

flighty_get_flight_stats

Aggregate statistics across your flight history.

Parameters:

  • year (integer, optional) — filter to a specific year; omit for all-time

Returns: Total flights, distance (km and miles), earth circumnavigations, unique departure/arrival airports, unique airlines, countries_visited (deduped across departure and arrival countries), cancelled flight count, top 5 airlines, top 5 routes.

Includes both auto-detected commercial flights and manually-entered flights (e.g., private/charter operators not in Flighty's commercial database).


flighty_get_connections

Layover information for connecting flight pairs. Sorted by first leg departure time descending.

No parameters.

Returns: For each connection: inbound flight (first leg, e.g., "AA1449"), origin airport, connection airport, outbound flight (second leg, e.g., "AA1166"), destination airport, arrival/departure times, layover duration in minutes, and minimum connection time.


flighty_list_friend_flights

Flights from your Flighty-connected friends. Same sort behavior as list_flights — upcoming sorted soonest first, otherwise most recent first.

Parameters:

  • friend_name (string, optional) — partial match on name
  • upcoming_only (boolean, default: false) — future flights only, sorted soonest first
  • year (integer, optional)
  • limit (integer, 1-200, default: 50)
  • offset (integer, default: 0)

flighty_add_flight

Adds a flight to your Flighty account via the Flighty API. The flight syncs to all your devices (phone, watch, widgets) within seconds.

Parameters:

  • flight_code (string) — e.g., "DL10", "UA194". The 2-character airline prefix is parsed automatically
  • date (string) — departure date in YYYY-MM-DD format

Returns: The server-side flight UUID on success, or an error if the flight isn't found for that date.


flighty_remove_flight

Permanently removes a flight from your Flighty account across all devices.

Parameters:

  • flight_id (string) — the flight UUID (from list_flights or add_flight results)

Warning: This cannot be undone. You would need to re-add the flight.


flighty_about

Returns server version, author, repository link, tool inventory, and requirements.

No parameters.

How it works

Flighty stores all flight data in a local SQLite database on macOS at:

~/Library/Containers/com.flightyapp.flighty/Data/Documents/MainFlightyDatabase.db

This server opens that database in read-only mode to answer queries. Queries UNION Flighty's Flight + ManualFlight and UserFlight + UserManualFlight tables so manually-entered flights surface alongside auto-detected commercial ones, and filter on UserFlight.isMyFlight = 1 so flights the user is following from friends (isMyFlight = 0) don't leak into "your flights" queries. It reads all necessary credentials directly from the installed Flighty app:

  • User identity — JWT auth token from Flighty.sqlite (identifies the user to the API)
  • API access — build token from the app's Info.plist (identifies the app version)
  • Sync token — from the app's UserDefaults plist (for flight deletion)

For write operations (add/remove flights), the server makes the same API calls the Flighty app itself makes. Added flights get full enrichment from Flighty's servers — gate assignments, weather, delay forecasts, codeshare data — and sync to all devices. No manual configuration is needed.

All tool responses include a note field with contextual metadata (sort order, timestamp format) to help AI assistants interpret results correctly. All timestamps are UTC.

Architecture

src/
  index.ts              # Entry point, tool registration, flighty_about
  constants.ts          # Database paths, API config, build token (read from app)
  types.ts              # TypeScript interfaces
  services/
    database.ts         # FlightyDatabase — all SQLite read queries
    flighty-api.ts      # FlightyApi — search, subscribe, delete via Flighty API
  tools/
    flights.ts          # list, current, get, search flights
    flight-status.ts    # status, delay forecast
    friends.ts          # friend flights
    reference.ts        # airport/airline search
    stats.ts            # aggregate statistics
    connections.ts      # layover/connection info
    write.ts            # add_flight, remove_flight (API-backed)

Limitations

  • macOS only — relies on the Flighty macOS app's sandboxed data
  • Flighty app must be installed and signed in — the server reads its local databases for both flight data and authentication
  • Read data freshness — flight status data is as fresh as the Flighty app's last sync, not real-time
  • Write operations require network — add/remove flights call the Flighty API
  • Flighty API is private — this server uses Flighty's undocumented API, which could change in future app updates
  • Build token is read from the installed app — updates automatically when the app updates, but if Flighty changes where it stores the token, the server will need updating

License

MIT

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured