Skip to main content
Glama

ThinAir Geo MCP Server

Give your AI agent complete geospatial awareness — geocode addresses and intersections, route across multiple modes (auto, truck, bicycle, pedestrian), check live traffic and weather, draw reachability polygons, and search places. Confidence-scored results so your agent knows when to ask again. 10 tools via MCP.

npm version License: MIT

What It Does

ThinAir Geo is a Model Context Protocol (MCP) server that gives AI agents full-stack geospatial tooling — no API integration required. Coordinates you can actually route from: intersection-level precision with confidence scores, no silent fallbacks (returns an empty result instead of a city-centroid guess when uncertain).

  • Geocoding — Forward, reverse, intersection, and batch (50/call). Results include confidence and method so agents can branch on quality.

  • Routing — Multi-profile: auto, truck, bicycle, pedestrian, motorcycle. 10 truck-specific presets (dry van, flatbed, step-deck, tanker, box truck, auto-carrier, sprinter, double trailer, oversize) with hazmat / dimensions / weight / time-distance constraints native.

  • Traffic — Real-time conditions for 30+ US metros.

  • Weather — Current conditions and forecasts.

  • Place Search & Explore — Nearby POIs, locality discovery with population filtering.

  • Isochrones — Drive-time / walk-time reachability polygons (multiple bands per call).

Tools

Tool

Description

geocode

Convert addresses, place names, or intersections to coordinates

reverse_geocode

Convert coordinates to addresses

batch_geocode

Geocode up to 50 addresses in one call (returns geojson FeatureCollection)

directions

Turn-by-turn routing — auto, truck, bicycle, pedestrian, motorcycle

isochrone

Travel-time / -distance reachability polygons (multi-band)

traffic

Real-time traffic conditions

weather

Current conditions and forecasts

search_places

Find nearby POIs by category (gas, charging, food, lodging, …)

explore

Browse cities, neighborhoods, and venues by region

quota

Inspect your current plan, daily usage, and remaining capacity (free, never debits)

Plus an issue_api_key utility — exposed via MCP for the API-key issuance flow but not part of the customer-facing 10-tool surface. Most clients complete OAuth at first use and never call it directly.

Truck Vehicle Profiles

Pass any of these as vehicle_profile on directions or isochrone to set dimensions automatically:

Profile

Dimensions / Use case

DRY_VAN_53

53′ dry van — standard US freight (default)

FLATBED_48

48′ flatbed, loaded height 14′

FLATBED_40

40′ flatbed, regional / specialty

STEP_DECK

Step-deck / drop-deck, 11′3″ upper-deck clearance

TANKER

Liquid/gas tanker (pair with hazmat=true)

BOX_TRUCK_26

26′ box truck, local delivery

AUTO_CARRIER

Car hauler / auto transport

SPRINTER_VAN

Cargo van, last-mile (9′ tall)

DOUBLE_TRAILER

Twin 28′ pups, LTL long-haul (92′ total)

OVERSIZE

Permitted oversize loads — override dimensions per haul

Individual params (truck_height, truck_weight, truck_length, axle_count, etc.) override the preset.

Quick Start

Add to your client's MCP config (e.g. ~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "thinair-geo": {
      "url": "https://geo.thinair.co/mcp"
    }
  }
}

The OAuth flow completes at first use — no manual token setup required. Free tier: 15 requests/day, no card.

API key (for non-OAuth clients)

{
  "mcpServers": {
    "thinair-geo": {
      "url": "https://geo.thinair.co/mcp",
      "headers": {
        "Authorization": "Bearer ta_live_..."
      }
    }
  }
}

Get a key at https://geo.thinair.co/connect.

npx (CLI / scripts)

npx -y @thinairtelematics/geo

Prints a config block to stdout for your client. Same OAuth-keyless option as above by default.

npm Package

npmjs.com/package/@thinairtelematics/geo

Previously published as thinair-geo (now deprecated in favor of the scoped package).

License

MIT © ThinAir Telematics

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ThinAirTelematics/thinair-geo'

If you have feedback or need assistance with the MCP directory API, please join our Discord server