Location Drive MCP Server

Connect Claude, Cursor, or any MCP-compatible AI client to the global Location Drive POI database — 350M+ places, graded building polygons, brand intelligence, and premium data packs.

Endpoint mcp.locationdrive.com/mcp npm @locationdrive/mcp 12 read-only tools

Hosted — no install

Streamable HTTP with your own API key as a Bearer header. Quotas and usage metering apply exactly as with direct API calls; every plan gets every tool and field. Get a free key →

Claude Code

claude mcp add --transport http locationdrive https://mcp.locationdrive.com/mcp \
  --header "Authorization: Bearer ld_live_YOUR_KEY"

Cursor — .cursor/mcp.json

{
  "mcpServers": {
    "locationdrive": {
      "url": "https://mcp.locationdrive.com/mcp",
      "headers": { "Authorization": "Bearer ld_live_YOUR_KEY" }
    }
  }
}

Local — runs on your machine

Requires Node 18+. Same tools, same key, over stdio.

Claude Desktop — claude_desktop_config.json

{
  "mcpServers": {
    "locationdrive": {
      "command": "npx",
      "args": ["-y", "@locationdrive/mcp"],
      "env": { "LOCATIONDRIVE_API_KEY": "ld_live_YOUR_KEY" }
    }
  }
}

Claude Code

claude mcp add locationdrive -e LOCATIONDRIVE_API_KEY=ld_live_YOUR_KEY \
  -- npx -y @locationdrive/mcp

Tools

ALL PLANSsearch_places

Search the global POI database by name or keyword — ranked by prominence for general terms, by similarity for specific names. Cost: limit per page (default 20); max_results auto-pages and sums the cost.

ALL PLANSfind_nearby

Distance-sorted places around a coordinate. Cost: limit per page (default 20); max_results auto-pages.

ALL PLANSbrand_locations

Every location of a brand — one country or global — with the total. Cost: limit per page (default 20); max_results auto-pages.

ALL PLANSget_place_details

Full record by ID; any of the 98 fields, premium packs (full menus, hotel rates, EV connectors, fuel, review text) included on every plan — with a data note when a place has none. Cost: 1.

ALL PLANSget_place_context

LLM-ready summary + structured record (sentiment keywords, not review text) for one place. Cost: 1.

ALL PLANSget_review_history

Merged multi-year review text history, duplicates removed, plus sentiment summary. Cost: 1.

ALL PLANSscan_reviews

Keyword scan of review text across up to 40 places near a point — matching snippets per place. Cost: the nearby search (its limit, default 20) + 1 per scanned place; results report api_calls_counted.

ALL PLANSget_hotel_rates

Compare room rates across booking sites — per-site offers with links. Cost: 1.

ALL PLANSget_building_polygon

GeoJSON footprint with A–E accuracy grade and area. Cost: 1.

ALL PLANSfind_building_at_point

Reverse lookup — which building contains this coordinate. Cost: 1.

ALL PLANSbrand_footprint

Store counts by country, open-now, coverage, average rating. Cost: 1 (summary).

ALL PLANSdata_coverage

Per-country POI counts and polygon coverage stats. Cost: 1.

Try asking

“How many Starbucks locations are in the dataset, and in which countries?”
“What restaurants are within 500 m of 40.7074, −74.0131?”
“Get the building footprint for this place and tell me its accuracy grade.”
“Which building is at 33.7144, −118.0058, and what businesses are inside?”
“Compare room rates for this hotel across booking sites.”
“How have this restaurant's reviews changed over the past few years?”
“Which cafés within 1.5 km of 30.0289, 31.4910 have reviews mentioning racism or hijab?”

Review text is per place: reviews, historical_reviews and get_review_history return what reviewers wrote (every plan). There is no city- or country-wide search over review content — scan_reviews covers a neighborhood (up to 40 places per call).

Usage: all plans receive all 98 fields. Usage is counted per place returned: list tools cost their limit (default 20); single-place and summary tools cost 1. Free keys: 5,000 API calls/month, 20 results per call. Every result carries api_calls_counted.

Test without quota: ld_test_ keys return synthetic data and never count against your plan — ideal for trying the integration before going live.