Leopaexd/stockholm-public-transport-mcp

SL Transit MCP Server

A Model Context Protocol (MCP) server providing Stockholm Local Transit (SL) journey planning and real-time information for daily commute monitoring and journey planning.

Features

• Coordinate-based operations: Find stops and plan journeys using latitude/longitude coordinates
• Real-time departures: Get upcoming departures from nearby stops
• Journey planning: Calculate routes between any two points
• Traffic monitoring: Check for current disruptions and service alerts
• Saved routes: Monitor pre-configured commute routes
• Committee integration: Designed for AI agent workflows

Installation

1. Install dependencies:

pip install -e .

2. Configure saved routes (optional): Copy sl_transit_config.json and customize with your locations.

API Access

No API keys required! The SL APIs are free to use without authentication. Simply install and run the server.

Usage

MCP Tools

find_nearby_stops

Find all transit stops within a radius of given coordinates.

await find_nearby_stops(
    lat=59.3293,
    lon=18.0686,
    radius_meters=500,
    transport_modes=["metro", "bus"]
)

get_departures_near_location

Get upcoming departures from all stops near coordinates.

await get_departures_near_location(
    lat=59.3293,
    lon=18.0686,
    radius_meters=300,
    transport_mode="metro",
    time_window_minutes=30
)

plan_journey_by_coordinates

Plan a journey between two coordinate points.

await plan_journey_by_coordinates(
    origin_lat=59.3293,
    origin_lon=18.0686,
    dest_lat=59.3326,
    dest_lon=18.0649,
    departure_time="2024-01-15T08:30:00",
    transport_modes=["metro", "bus"]
)

get_deviations

Get current traffic disruptions and service alerts.

await get_deviations(
    transport_modes=["metro"],
    lines=["Red line"],
    severity="high"
)

check_saved_route

Check status of a pre-configured route.

await check_saved_route("home_to_work")

Configuration

The server loads configuration from sl_transit_config.json:

{
  "saved_routes": {
    "home_to_work": {
      "name": "Home to Work",
      "origin": {
        "lat": 59.3293,
        "lon": 18.0686,
        "label": "Home"
      },
      "destination": {
        "lat": 59.3326,
        "lon": 18.0649,
        "label": "Office"
      },
      "typical_duration_minutes": 35,
      "notify_if_delay_exceeds_minutes": 5,
      "preferred_transport_modes": ["metro", "bus"]
    }
  },
  "favorite_locations": {
    "home": {
      "lat": 59.3293,
      "lon": 18.0686,
      "label": "Home"
    }
  }
}

Committee Integration

This server is designed to work with AI planning agents (the "Oliver Care Committee"):

Morning Planning

# Check commute before daily planning
commute_status = await check_saved_route("home_to_work")
if "delay" in commute_status:
    # Add note to daily plan
    # Suggest earlier departure

Afternoon Reminders

# Help Oliver leave on time
work_to_home = await check_saved_route("work_to_home")
# Calculate when to leave
# Provide ADHD-friendly reminder

Meeting Planning

# Calculate travel time for external meetings
journey = await plan_journey_by_coordinates(
    origin_lat=office_lat,
    origin_lon=office_lon,
    dest_lat=meeting_lat,
    dest_lon=meeting_lon
)
# Add buffer time for delays
# Include in meeting notes

Development

Run the server:

python server.py

Run tests:

pytest

Format code:

black server.py
ruff check server.py

Quick Start

Since no API keys are required, you can start using the server immediately:

1. Install: pip install -e .
2. Run: python server.py
3. Test with coordinates: Use any Stockholm area coordinates to find nearby stops

Logging

Logs are written to logs/sl_transit_mcp.log with weekly rotation and monthly retention.

License

MIT License - see LICENSE file for details.