Skip to content

search-along-route.md

Latest commit

 

History

History
233 lines (172 loc) · 4.75 KB

search-along-route.md

File metadata and controls

233 lines (172 loc) · 4.75 KB
Raw

Search Along Route Examples

This document provides examples for testing the search-along-route prompt in the MCP Inspector or any MCP client.

Basic Usage

Example 1: Coffee Shops on a Road Trip

Prompt Arguments:

{
  "from": "Seattle, WA",
  "to": "Portland, OR",
  "search_for": "Starbucks"
}

Expected behavior:

  • Geocodes Seattle and Portland
  • Gets driving route (~175 miles)
  • Creates 1km buffer corridor on each side of I-5
  • Searches for Starbucks locations
  • Filters to corridor, orders by route progress
  • Displays map with route line and coffee shop markers

Example 2: Gas Stations on Highway Route

Prompt Arguments:

{
  "from": "Los Angeles, CA",
  "to": "San Francisco, CA",
  "search_for": "gas stations",
  "mode": "driving",
  "buffer_meters": "2000"
}

Expected behavior:

  • Gets driving route along I-5/US-101 (~380 miles)
  • Creates 2km buffer (wider for highway)
  • Finds all gas stations within corridor
  • Orders by position along route
  • Shows results with distance markers

Example 3: Rest Stops on Long Drive

Prompt Arguments:

{
  "from": "Denver, CO",
  "to": "Salt Lake City, UT",
  "search_for": "rest stops",
  "mode": "driving"
}

Expected behavior:

  • Gets I-70 route (~520 miles)
  • Searches for rest areas and travel plazas
  • Shows results ordered by route progress

Example 4: Restaurants on Walking Route

Prompt Arguments:

{
  "from": "Times Square, New York, NY",
  "to": "Central Park, New York, NY",
  "search_for": "restaurants",
  "mode": "walking",
  "buffer_meters": "500"
}

Expected behavior:

  • Gets walking route
  • Creates 500m buffer (narrower for walking)
  • Finds restaurants within walking corridor
  • Shows results ordered by route position

Example 5: Bike Shops on Cycling Route

Prompt Arguments:

{
  "from": "Golden Gate Bridge, San Francisco",
  "to": "Sausalito, CA",
  "search_for": "bike shops",
  "mode": "cycling",
  "buffer_meters": "750"
}

Expected behavior:

  • Gets cycling route across bridge
  • Searches for bike shops and repair stations
  • Orders by route progress
  • Shows on map with cycling-friendly visualization

Testing in MCP Inspector

  1. Start the inspector:

    npm run inspect:build

  2. Navigate to Prompts section

  3. Select search-along-route prompt

  4. Fill in arguments:

    • from: Seattle, WA
    • to: Portland, OR
    • search_for: Starbucks
    • (leave mode and buffer_meters as defaults)

  5. Click "Run Prompt"

  6. Expected output: The AI will receive instructions to:

    • Geocode both locations
    • Get the route using directions_tool
    • Create buffer using buffer_tool
    • Search using category_search_tool or search_and_geocode_tool
    • Filter using points_within_polygon_tool
    • Order using distance_tool
    • Visualize on map

Advanced Examples

Multiple Segments (Long Routes)

For very long routes (>100km), the prompt suggests searching in segments:

Prompt Arguments:

{
  "from": "Boston, MA",
  "to": "Miami, FL",
  "search_for": "rest stops",
  "mode": "driving",
  "buffer_meters": "3000"
}

Expected behavior:

  • Route is ~1,500 miles
  • Prompt instructs to search in segments
  • Avoids overwhelming API with single large bbox
  • Results grouped by route section

Adjusting Buffer for Different Contexts

Urban area (narrow):

{
  "from": "Union Station, Chicago",
  "to": "Navy Pier, Chicago",
  "search_for": "parking garages",
  "mode": "driving",
  "buffer_meters": "250"
}

Highway (wide):

{
  "from": "Phoenix, AZ",
  "to": "Las Vegas, NV",
  "search_for": "truck stops",
  "mode": "driving",
  "buffer_meters": "5000"
}

Verification Checklist

When testing, verify the prompt output includes:

  • ✅ Geocoding instructions (if needed)
  • ✅ Route retrieval with correct travel mode
  • ✅ Buffer creation with specified distance
  • ✅ Search within bounding box
  • ✅ Point-in-polygon filtering
  • ✅ Distance calculation and ordering
  • ✅ Map visualization instructions
  • ✅ Result formatting (name, address, distance, route position)
  • ✅ Contextual information (route distance, travel time)
  • ✅ Helpful notes (adjust buffer if no results, segment for long routes)

Common Use Cases

  • Road trip planning: Find chain restaurants along highway routes
  • EV charging: Locate charging stations on long drives
  • Fuel planning: Find gas stations with optimal spacing
  • Amenities: Rest stops, hotels, restaurants on travel routes
  • Emergency services: Hospitals, urgent care along routes
  • Tourist stops: Attractions, viewpoints along scenic routes
  • Delivery logistics: Pickup/dropoff locations near planned routes