MCP server

GeoSpark

io.github.Maz2580/geospark

Checked 25 Oct 2026, 20:29 UTC · Connected to it and listed its tools

Verifiable geospatial tools: distances, topology, isochrones, Overture Maps, with sources.

Website Source

At a glance

Type

MCP server

Publisher

Not verified

Status

Working at last check

Tools

21

Token cost

About 5,353 tokens per request

Sets up in

Claude Code, Claude Desktop, Codex, Gemini CLI, Cursor, VS Code, Antigravity, ChatGPT, OpenAI API, Grok

Last checked

25 Oct 2026, 20:29 UTC

Connect it to your agent

Add to CursorAdd to VS CodeVS Code Insiders

Opens the app, which shows the server and asks you to confirm. For other apps, copy the setup below.

Claude Code

claude mcp add --transport http geospark https://geospark.terrascout.app/mcp

Claude Desktop

Settings → Connectors → Add custom connector, and paste the URL.

Codex (config.toml)

[mcp_servers.geospark]
url = "https://geospark.terrascout.app/mcp"

Gemini CLI

{"mcpServers":{"geospark":{"httpUrl":"https://geospark.terrascout.app/mcp"}}}

Cursor

{"mcpServers":{"geospark":{"url":"https://geospark.terrascout.app/mcp"}}}

VS Code

{"servers":{"geospark":{"type":"http","url":"https://geospark.terrascout.app/mcp"}}}

Antigravity

agy mcp add geospark https://geospark.terrascout.app/mcp

ChatGPT

Settings → Apps & Connectors → Advanced settings: turn on Developer mode, then Create a connector and paste the URL. Which plans can do this varies.

OpenAI API (Responses tools)

{"type":"mcp","server_label":"geospark","server_url":"https://geospark.terrascout.app/mcp","require_approval":"always"}

Grok (xAI API tools)

{"type":"mcp","server_label":"geospark","server_url":"https://geospark.terrascout.app/mcp"}

Tools

• calculate_distance ~436 tokens
  Exact geodesic distance between two places or geometries on the WGS84 ellipsoid. Use for any "how far" question instead of estimating. For NAMED places pass place_a and place_b; the server geocodes them, so never type coordinates for a place yourself. Use point_a/point_b onl
• check_spatial_relationship ~266 tokens
  Ground-truth topology: does geometry_a <relationship> geometry_b? Use instead of guessing containment, overlap or adjacency. Invalid geometries (for example self-intersecting polygons) are refused with the reason.
• convert_height ~336 tokens
  Convert a height between the ellipsoid (GNSS, drones) and a national height datum with the official geoid model: AHD (AUSGeoid2020), ODN (OSGM15), NZVD2016 (NZGeoid2016), NAVD88 (GEOID18), or global EGM2008. Never estimate a geoid separation from memory: it is often tens of m
• geocode ~157 tokens
  Find coordinates for an address or place name (OpenStreetMap Nominatim). Check `geospark.warnings` for ambiguous names (e.g. "Springfield") and ask the user which one they mean, or add a region.
• get_active_fires ~165 tokens
  Satellite active-fire detections near a point (NASA FIRMS; the server needs a FIRMS_MAP_KEY).
• get_air_quality ~158 tokens
  Latest PM2.5, NO2, O3 from the nearest monitoring stations (OpenAQ; the server needs an OPENAQ_API_KEY), with each station's data licence.
• get_elevation ~149 tokens
  Ground elevation in metres (Copernicus GLO-30, Open Elevation fallback), with its vertical datum and accuracy.
• get_weather ~163 tokens
  Current weather and a daily forecast for a point (Open-Meteo, no key needed).
• isochrone ~230 tokens
  Where can you get to from a point within N minutes on foot, by bike or by car? Returns one polygon per contour with its geodesic area (Valhalla over OpenStreetMap).
• line_of_sight ~391 tokens
  Whether A and B can see each other over the terrain (Copernicus 30 m DEM, curvature and refraction), the minimum clearance and where it occurs; with frequency_mhz, the first Fresnel zone clearance of a radio link (60% clear is the usual rule).
• location_report ~153 tokens
  A dossier for a place: address, elevation, weather, and nearby amenities with exact distances, as a GeoJSON map layer listing every data source used. Takes ~10-30 s.
• overture_features ~382 tokens
  Open global map data from Overture Maps, no API key: points of interest (with categories), building footprints (height, floors, geodesic footprint area), road and path segments (geodesic length), or `divisions`, the administrative areas containing the point (country, region,
• reverse_geocode ~136 tokens
  Find the address or place at a coordinate (OpenStreetMap Nominatim).
• route ~236 tokens
  Driving route between two points (OSRM over OpenStreetMap): line geometry, distance in metres and duration in seconds.
• select_site ~253 tokens
  Rank candidate sites in an area by geodesic distance to amenities you want nearby and away from ones to avoid (OpenStreetMap data). Takes ~10-30 s.
• spatial_operation ~227 tokens
  Compute a buffer (shape-preserving, metres), geodesic area or length, centroid or convex hull. Returns the resulting geometry as GeoJSON with its measurements and units.
• sun_position ~317 tokens
  The Sun's apparent elevation and azimuth (from true north) at a place and time, and the shadow a vertical object casts on level ground (NOAA solar position, refraction included).
• sun_times ~282 tokens
  Sunrise, solar noon, sunset and day length on a date for a flat horizon; with terrain=True, also the hours of direct sun once hills are taken into account.
• terrain_horizon ~312 tokens
  The skyline the terrain makes around a point: for each azimuth, the elevation angle of the horizon and how far away the blocking terrain is (Copernicus 30 m DEM, Earth curvature and refraction). Use for solar-panel shading, valley sunlight, or what hills hide. ~10-60 s.
• transform_coordinates ~366 tokens
  Convert coordinates between CRSs as the official transformation does: GDA94 to GDA2020, MGA zones, British National Grid (OSTN15), NAD27 to NAD83 (NADCON), ITRF at an epoch, and any other EPSG pair. Uses national grids (downloaded once), refuses PROJ's ballpark approximations
• travel_time_matrix ~238 tokens
  Network travel time (minutes) and distance (metres) from each origin to each destination. Note: points here are [latitude, longitude] pairs.

Browse

• GitHub tools
• Hiring and jobs
• GitHub and code hosting
• Weather
• Maps and places

Related MCP servers

• Geobble
  Geospatial workspace: manage sources and maps, run spatial analysis, and publish interactive maps
• Mapsource OpenStreetMap & Geospatial APIs
  OpenStreetMap queries, maps/styles, search, routing, terrain, analysis, pipelines, and rendering.
• Geoapify MCP by usefulapi
  Geocode, reverse-geocode, autocomplete, route and search places.
• Jean — Molecular Biology & Genomics
  Validated molecular biology tools: cloning, primers, CRISPR, Sanger, alignment, qPCR, phylogenetics
• HubVibe: Pay-per-Call Agent Tools: Web Search, Email Verify, KYC Screening, Stocks, Crypto, News
  70+ pay-per-call tools for AI agents: web search, email verify, KYC, stocks, crypto, news, data
• MapMap
  Maps for agents: truck/ADR routing, geocoding, places, matrices, isochrones, EV and fleet planning.
• Topology Index
  Reference for multi-agent LLM topologies: patterns, aliases, sources, starters; one gated report.
• Diogenes Autonomous Microservice Gateway
  Autonomous AI microservices: 55 dynamic guardrails, contract audits & oracles via L402/x402/MPP.

Badge for your README

[![AptLynx](https://aptlynx.net/badge/io.github.Maz2580/geospark.svg)](https://aptlynx.net/servers/io.github.Maz2580/geospark)

Listing text is written by the maker; AptLynx lists tools and does not vouch for them. Find another tool · Browse by topic · For agents