MCP Server for Location Intelligence - Claude & ChatGPT Integration

What is MCP?

Model Context Protocol (MCP) is an open protocol that allows AI applications (like Claude Desktop) to connect to external data sources and tools.

Camino AI's MCP server enables AI agents to:

• Search for locations using natural language
• Understand queries like "find hotels near the Eiffel Tower"
• Get AI-ranked, contextually relevant results
• Access location intelligence without writing code

Remote MCP Server

Unlike most MCP servers that require local installation, Camino AI uses a remote-first approach similar to Tavily:

• ✅ Zero local installation - No packages to install or manage
• ✅ Always up-to-date - Automatic updates on our server
• ✅ Works everywhere - Cloud-hosted, accessible from anywhere
• ✅ One-line setup - Just add configuration and your API key

MCP Server URL:

https://api.getcamino.ai/mcp?caminoApiKey=YOUR_API_KEY

Claude Desktop Integration

Setup Instructions

1. Get your API key from app.getcamino.ai/dashboard

2. Open Claude Desktop config file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

3. Add Camino AI MCP server:

{
  "mcpServers": {
    "camino-ai": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-http",
        "https://api.getcamino.ai/mcp?caminoApiKey=YOUR_API_KEY_HERE"
      ]
    }
  }
}

4. Restart Claude Desktop

5. Verify installation: Look for the hammer/tool icon (🔨) in Claude Desktop. You should see the camino_query tool available.

Cursor IDE Integration

Same configuration works for Cursor IDE:

// Add to Cursor MCP config
{
  "mcpServers": {
    "camino-ai": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-http",
        "https://api.getcamino.ai/mcp?caminoApiKey=YOUR_KEY"
      ]
    }
  }
}

Available MCP Tools

camino_query

Search for places using natural language with AI-powered ranking.

Parameters:

Usage Examples

Basic Location Search

User to Claude: "Find coffee shops near Times Square in New York"

Claude uses MCP tool:

{
  "name": "camino_query",
  "arguments": {
    "query": "coffee shops near Times Square New York",
    "latitude": 40.7589,
    "longitude": -73.9851,
    "limit": 10
  }
}

Claude returns: "I found 10 coffee shops near Times Square. The closest is Starbucks Reserve (0.2 miles away) at 1271 6th Avenue, followed by Blue Bottle Coffee (0.3 miles) known for their pour-over coffee..."

Contextual Search

User: "I need a quiet place to work in San Francisco, preferably with good WiFi"

Claude uses MCP:

{
  "name": "camino_query",
  "arguments": {
    "query": "quiet coffee shops with WiFi good for working",
    "latitude": 37.7749,
    "longitude": -122.4194,
    "rank": true,
    "generate_answer": true
  }
}

Travel Planning

User: "What are the top museums to visit in Paris?"

Claude uses MCP:

{
  "name": "camino_query",
  "arguments": {
    "query": "famous museums and art galleries in Paris",
    "latitude": 48.8566,
    "longitude": 2.3522,
    "radius": 5000,
    "rank": true
  }
}

No Coordinates Provided

User: "Find hotels in Tokyo"

Claude uses MCP (AI generates coordinates):

{
  "name": "camino_query",
  "arguments": {
    "query": "hotels in Tokyo",
    "limit": 15
  }
}

Note: Camino AI automatically generates approximate coordinates (~35.6762, 139.6503) for "Tokyo"

MCP Endpoints

JSON-RPC Protocol (Primary)

POST https://api.getcamino.ai/mcp/rpc?caminoApiKey=YOUR_KEY
Content-Type: application/json

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "camino_query",
    "arguments": {
      "query": "restaurants in Times Square",
      "latitude": 40.7589,
      "longitude": -73.9851,
      "limit": 5
    }
  }
}

REST Endpoints (Legacy)

GET https://api.getcamino.ai/mcp/tools?caminoApiKey=YOUR_KEY
POST https://api.getcamino.ai/mcp/tools/call?caminoApiKey=YOUR_KEY

Real-World Use Cases

Authentication & Security

• API Key Authentication: API key passed via URL parameter caminoApiKey
• Real-time Validation: Keys validated against Firestore on each request
• Rate Limiting: 30 requests per minute per API key
• Usage Tracking: All MCP calls count toward monthly API quota
• Secure Storage: Store API key in config file, never in code or version control

Troubleshooting MCP Integration

Tool not appearing in Claude Desktop

• Verify config file is in correct location
• Check JSON syntax is valid (use a JSON validator)
• Ensure API key is correct (no extra spaces)
• Restart Claude Desktop completely
• Check Claude Desktop logs for errors

Tool calls failing

• Verify API key is active in dashboard
• Check you haven't exceeded rate limits (30/min)
• Ensure you haven't hit monthly usage limit
• Try with simpler query to isolate issue

Slow responses

• AI ranking adds ~200-400ms processing time
• Set rank: false for faster responses
• Reduce limit parameter for fewer results
• Check your internet connection

MCP vs Direct API

MCP Pricing

MCP calls use the same pricing as direct API calls:

All MCP tool calls count toward your monthly API usage quota.

Additional Resources

• Model Context Protocol Specification
• Camino AI MCP Server Info
• Available MCP Tools List
• Camino AI API Documentation