AI Skill
SKILL.md

Domain knowledge for London Transit — workflow patterns, data models, and gotchas for your AI agent.

London Transit (Transport for London Unified API)

Real-time data for London's entire public transport network. Powered by TfL Open Data.

Interactive Map App

An interactive map dashboard is available alongside this chat. When the app panel is open, your tool results are automatically forwarded to the map — do NOT call the app-open tool again. Just use the data tools below (get status, plan journey, search stops, etc.) and the map will update with the results rendered as map layers: tube lines as colored polylines, stations as markers, bike docks as colored circles, arrivals as floating boards.

Data Model

  • Lines — Transport lines across modes: tube, dlr, overground, elizabeth-line, bus, tram, cable-car, river-bus.
  • StopPoints — Stations, bus stops, piers. Identified by NaPTAN IDs (e.g. 940GZZLUVIC for Victoria Tube) or ICS codes (e.g. 1000248).
  • Predictions — Live arrival/departure predictions per line or stop. timeToStation is in seconds.
  • Journeys — Planned routes between two points with legs, duration, and fare information.
  • BikePoints — Santander Cycles docking stations with live bike/empty dock counts.

Key Patterns

  • Tube line IDs are lowercase: bakerloo, central, circle, district, hammersmith-city, jubilee, metropolitan, northern, piccadilly, victoria, waterloo-city, elizabeth.
  • Journey planner works best with ICS codes (the ics_id field from search results, e.g. 1000108 for Highbury & Islington). Also accepts NaPTAN IDs, Hub IDs (auto-resolved), or coordinates (lat,lon format as a string like 51.5074,-0.1278).
  • StopPoint search returns matches with both id (Hub/NaPTAN) and ics_id (ICS code) fields. Use ics_id for journey planning.
  • BikePoint additional properties contain NbBikes, NbEmptyDocks, NbDocks.
  • Line status severity: 10=Good Service, 9=Minor Delays, 6=Severe Delays, 5=Part Closure, 4=Planned Closure, 1=Part Suspended, 0=Suspended, 20=Service Closed.

Gotchas

  • Arrivals are unordered: The /Arrivals endpoint returns predictions in arbitrary order — sorted by timeToStation in tool output.
  • Journey planner station IDs: Use ICS codes (e.g. 1000266 for Westminster) or NaPTAN IDs (940GZZLUWSM). Full stop names don't work — use search stops first.
  • BikePoint full list is large: ~800 stations. Always search by area name.
  • Mode names: Use lowercase with hyphens: tube, dlr, overground, elizabeth-line, bus, tram, cable-car, river-bus.

Where the data comes from

Tool calls are proxied directly to the TfL Unified API with light normalization (sorted arrivals, resolved Hub/ICS/NaPTAN IDs, filtered modes). No caching, no transformation of source values.

Transport for London — Unified API
Transport for London (TfL)

Live operational data for London's public transport network: Tube/DLR/Overground/Elizabeth line statuses and arrivals, bus routes and live arrivals, journey planner, stop and station metadata, and Santander Cycles docking-station availability.

London-wide network — ~800 bike docks, ~20K stops, all TfL-operated lines
Source updated: Real-time
We refresh: Real-time (proxied per request, no caching)
JSON HTTP API
Greater London, UK
OpenStreetMap (basemap tiles)
OpenStreetMap contributors

Basemap tiles rendered by CARTO for the interactive map UI. Not used for any data lookup — display only.

Source updated: Continuous (community-edited)
We refresh: On client request (tiles served by carto CDN)
Raster PNG tiles
Global
Powered by TfL Open Data. Contains OS data © Crown copyright and database rights 2016, and Geomni UK Map data © and database rights 2019.
Live arrival predictions reflect TfL's best estimate at the time of the request and can change without notice. Do not use for safety-critical decisions.

Source availability last verified 2026-04-20.

Tools in this Server (7)

London Transit Bike Points

Search for Santander Cycles (bike-share) docking stations in London by name or area. Returns station names, locations, and IDs. Use a search term to f...

London Transit Journey Plan

Plan a journey between two points on the London transport network. Returns route options with duration, legs, line changes, departure/arrival times, a...

London Transit Line Arrivals

Get live arrival predictions for a specific TfL line, including London bus routes. Returns upcoming arrivals with station/stop names, platform, direct...

London Transit Line Status

Get the current service status of one or more specific TfL lines, including London bus routes. Returns status severity, description, and disruption re...

London Transit Search Stops

Search for TfL stops and stations by name. Returns matching stops with their IDs (NaPTAN/ICS codes usable in journey planning), transport modes, zones...

London Transit Stop Arrivals

Get live arrival predictions for a specific stop or station. Returns upcoming arrivals across all lines serving that stop, with platform, direction, d...

London Transit Tube Status

Get the current service status of all London Underground (Tube) lines, or any other TfL mode (DLR, Overground, Elizabeth line, bus, tram, etc.). Retur...

Frequently Asked Questions

What is the London Transit MCP server?

Real-time London transport data — Tube line statuses, journey planning, stop search, live arrival predictions, and Santander Cycles docking stations. Covers Tube, DLR, Overground, Elizabeth line, buses, trams, and river bus. Powered by TfL Open Data. It provides 7 tools that AI agents can use through the Model Context Protocol (MCP).

How do I connect London Transit to my AI agent?

Add the MCPBundles server URL to your MCP client configuration (Claude Desktop, Cursor, VS Code, etc.). The URL format is: https://mcp.mcpbundles.com/bundle/london-transit. Authentication is handled automatically.

How many tools does London Transit provide?

London Transit provides 7 tools that can be called by AI agents, along with a SKILL.md that gives your AI agent domain knowledge about when and how to use them.

What authentication does London Transit require?

London Transit uses open data APIs — no authentication required.

Setup Instructions

Connect London Transit to any MCP client in minutes

https://mcp.mcpbundles.com/bundle/london-transit

What is MCP?

Model Context Protocol lets AI tools call external capabilities securely through a single URL. This bundle groups tools behind an MCP endpoint that many clients can use.

Use this bundle in 3 steps

  1. Copy the MCP URL above
  2. Open your AI tool and add a new MCP/connector
  3. Paste the URL and follow any auth prompts

Claude Desktop Users

Skip the manual setup! Use the .mcpb file format for one-click installation. Check the Claude Desktop tab for setup instructions.

Pick your tool tab for exact steps

Select ChatGPT, Cursor, Claude Code, or another tab for copy-paste config.

Other ways to use London Transit

Same data, different audiences.

App

Polished interactive UI — explore the data visually with no setup.

Open the app

Try London Transit now

No sign-up required. Open the interactive dashboard and start searching in seconds.