Directions from A to B, by name.

Geocode both ends and compute the route in one call. The agent says 'how do I get from Kings Cross to Tower Bridge'. The tool handles the rest.

mcp.footstep.ai / find_and_route
Why you need it

Why your agent needs this.

When the user says 'directions from Kings Cross to Tower Bridge', the agent shouldn't need to know it has to geocode first. find_and_route is the agent-native tool: pass two strings (or coordinates, or any mix), get back the route plus the matched start and end. The agent never orchestrates a multi-step flow.

This tool bundles Geocoding + Routing in a single call.

What it does for you

What your agent gets back.

  • Origin and destination accept lat/lon OR a place name, postcode, or free-text address.
  • Single tool call, single render envelope: LineString for the route plus Point layers for the matched ends.
  • Same terrain block as get_directions, ready for 'is this hilly?' follow-ups.
  • Reach for this on 'route from A to B' / 'directions from A to B' / 'take me from A to B' asks.
  • Composes with compare_routes when the user wants mode comparison too.
Try it

Plug it into your agent.

Add Footstep to your MCP-aware runtime. Your model picks the tool at runtime; you wrote none of the orchestration.

Prompt that triggers find_and_route
"Directions to the nearest train station from here."
claude_desktop_config.json
{
  "mcpServers": {
    "footstep": {
      "url": "https://mcp.footstep.ai",
      "headers": {
        "x-api-key": "sk_live_your_key_here"
      }
    }
  }
}

Ready to call it?

£5 free credit on signup. One auth header, every major runtime.