Skyfly

Skyfly MCP Server bridges live flight data with FAA aircraft references to empower context-rich AI queries. It lets you query real-time OpenSky data alongside a local FAA reference repository to answer complex aviation questions with up-to-date positions, trajectories, and aircraft specs.

How to use

You connect your MCP client to Skyfly to run intelligent queries that combine live flight data with aircraft specifications. Use the HTTP endpoint for remote access and the local MCP command for offline development. You can ask for current positions, regional aircraft lists, arrivals and departures, or enrich live results with model and engine details from the FAA references.

How to install

Prerequisites: you need Python 3.10 or newer and an SSL certificate (Let's Encrypt recommended). Then follow these steps to install and run Skyfly MCP Server.

# Clone the project
git clone https://github.com/vog01r/skyfly-mcp.git
cd skyfly-mcp

# Create and activate a virtual environment
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Download FAA data (optional but recommended)
# Place ACFTREF.txt, ENGINE.txt, MASTER.txt into the ReleasableAircraft folder
mkdir -p ReleasableAircraft

# Ingest FAA data
python -c "from aircraftdb.database import get_database; from aircraftdb.ingest import ingest_directory; from pathlib import Path; db = get_database(); ingest_directory(Path('ReleasableAircraft'), db)"

# Start the MCP server
./start.sh

Additional configuration and start commands

There are two ways to connect Skyfly from an MCP client. You can use the remote HTTP endpoint or run the server locally via the provided startup script.

Remote MCP URL (HTTP) example shown in the client configuration:

Security and deployment notes

Secure your deployment with HTTPS and a valid SSL certificate. If you are deploying behind a reverse proxy, configure your proxy to forward requests to the Skyfly MCP server as described in the production deployment steps.

Examples of common tasks

Query current aircraft in a region, fetch arrivals by an airport, or enrich a list of aircraft with FAA specs using the available tools.

Troubleshooting

If you encounter issues starting the server, verify that the Python environment is activated, dependencies are installed, and FAA data ingestion completed successfully. Check that the start script has execute permissions and that port access is allowed by your firewall.

Notes on data sources

Skyfly MCP combines real-time data from OpenSky Network with a local FAA reference repository that contains aircraft models, US registrations, and engine references for rich contextual queries.

Available tools

get_aircraft_states

Fetches current positions of aircraft in real time from OpenSky Network.

get_aircraft_in_region

Returns aircraft currently in a specified geographic region (e.g., France, Europe).

get_arrivals_by_airport

Provides arrivals for a given airport, based on real-time and historical data.

get_departures_by_airport

Provides departures for a given airport, based on real-time and historical data.

get_track_by_aircraft

Retrieves the trajectory history for a specific aircraft.

db_lookup_by_mode_s

Enriches an ICAO24 code with aircraft specifications from the FAA reference data.

db_search_aircraft

Searches the FAA aircraft registry for specific registrations or identifiers.

db_search_models

Searches aircraft models (e.g., Boeing, Cessna) in the FAA reference data.

db_enrich_live_aircraft

Enriches a list of ICAO24 codes with additional model and engine information from live FAA data.

db_sql_query

Executes a custom SQL query against the FAA reference database.