MCP Quickstart

Documentation Index

Fetch the complete documentation index at: https://docs.footstep.ai/llms.txt

Use this file to discover all available pages before exploring further.

​

What you can do

​

Plan a journey

• “How do I walk from Kings Cross to Tower Bridge?”
• “Directions to Buckingham Palace from my hotel”
• “What’s the fastest way to drive from Heathrow to Cambridge?”

​

Compare travel options

• “Should I walk or cycle from the office to the station?”
• “Compare driving vs cycling - which is hillier?”
• “What’s the flattest route between these two places?”

​

Find places

• “Find coffee shops near Kings Cross”
• “Are there any supermarkets within walking distance?”
• “What restaurants are near 51.53, -0.12?”

​

Find stops on the way

• “Are there any petrol stations on my route to Cambridge?”
• “Find a coffee shop on the way that won’t add more than 5 minutes”

​

Check reachability

• “How far can I cycle in 20 minutes from Liverpool Street?”
• “Show me everywhere I can walk to in 15 minutes from here”

​

Work with addresses

• “Where is 10 Downing Street?”
• “What address is at 51.5034, -0.1276?”
• “Geocode these 5 addresses and tell me which ones are in London”

​

Analyse terrain and elevation

• “What’s the elevation profile between these two coordinates?”
• “How much climbing is there on this walking route?”
• “Is this route flat or hilly?”

​

Analyse a recorded activity

• “Clean up this GPS trace from my bike ride”
• “What roads did I travel on during this run?”
• “Show me the terrain and surface quality of my recorded walk”

​

Optimise multiple stops

• “Find the best order to visit these 8 delivery addresses”
• “I need to visit 5 clients today - what’s the most efficient route?”

​

How to set it up

​

1. Get an API key

​

2. Connect your assistant

{
  "mcpServers": {
    "footstep": {
      "url": "https://mcp.footstep.ai",
      "headers": {
        "x-api-key": "sk_live_your_key_here"
      }
    }
  }
}

{
  "mcpServers": {
    "footstep": {
      "url": "https://mcp.footstep.ai",
      "headers": {
        "x-api-key": "sk_live_your_key_here"
      }
    }
  }
}

​

3. Start asking

​

Tips

• Be specific about how you want to travel. “Walking directions” and “cycling directions” give different results - cycling accounts for bike lanes and road surfaces, walking accounts for footpaths and stairs.
• Mention terrain if you care about it. Ask “what’s the flattest route?” or “how hilly is this?” and the assistant will include elevation and difficulty information.
• Give coordinates if you have them. You can mix place names and coordinates: “Directions from 51.53, -0.12 to Tower Bridge”.
• Ask follow-up questions. After getting directions, you can ask “is there a flatter alternative?” or “how about by bike instead?”.

​

Map-rendering clients

const result = await tools.get_directions.execute({
  locations: [{ lat: 51.5, lon: -0.1 }, { lat: 55.95, lon: -3.19 }],
});

for (const layer of result.render ?? []) {
  // layer.kind: "linestring" | "point" | "polygon" | "h3-cells" | …
  // layer.data: GeoJSON Feature / FeatureCollection (or kind-specific shape)
  // layer.bbox: [minLng, minLat, maxLng, maxLat]
  myMap.addLayer(layer);
}

​

Token-conscious LLM agents

import { experimental_createMCPClient } from "ai";

const mcpClient = await experimental_createMCPClient({
  transport: {
    type: "sse",
    url: "https://mcp.footstep.ai",
    headers: { "x-api-key": process.env.FOOTSTEP_API_KEY! },
  },
});

const rawTools = await mcpClient.tools();

// Set `compact: true` on every tool call to receive summary-only responses.
const tools = Object.fromEntries(
  Object.entries(rawTools).map(([name, tool]) => [
    name,
    {
      ...tool,
      execute: (args: Record<string, unknown>, opts: unknown) =>
        tool.execute({ ...args, compact: true }, opts),
    },
  ]),
);