← All BumiAtlas MCP endpoints

BumiAtlas MCP

Mongolia: eGazar land parcels, land use and addresses

Mongolian land parcels from ALAMGaC's eGazar cadastre, as tools an AI assistant can call: find a parcel by id, address or coordinates, read its land-use class, area and address, and see it on a map with its neighbours.

Server URL

Streamable HTTP with OAuth. The first time you connect, your client opens a browser window to sign in with Google; no API key needed.

https://bumiatlas.com/mcp/mongolia

Connect

Claude (claude.ai, desktop)
Settings → Connectors → Add custom connector, paste the URL, then Connect and sign in with Google.
Claude Code
claude mcp add --transport http bumiatlas-mongolia https://bumiatlas.com/mcp/mongolia
ChatGPT
Settings → Apps & Connectors → Advanced → Developer mode, then Create with the URL and OAuth authentication.
Cursor
in ~/.cursor/mcp.json
{
  "mcpServers": {
    "bumiatlas-mongolia": {
      "url": "https://bumiatlas.com/mcp/mongolia"
    }
  }
}

Tools

Tool names and descriptions are in English, which models understand best; your assistant still answers in your language.

MN_parcel_search
Find Mongolian land parcels (ALAMGaC eGazar) by parcel id, address or location, with each parcel's whole record. query takes a 10-digit parcel id (also matches old ids from before a split or merge), coordinates as 'latitude, longitude' (e.g. 47.9186, 106.9177), or address words in Mongolian Cyrillic: street or neighbourhood names and the khashaa number (e.g. 'Билүүт 106'); every word must match. Narrow an address with au2 (a district code or its prefix; 04 = Ulaanbaatar). With coordinates it returns parcels containing the point (distance 0), then the nearest within `radius` metres, closest first. Each parcel has id, oldParcelId, landuse (class code), landuseName (Mongolian), landuseEn, area (m², as recorded), street, khashaa, neighbourhood, au2, validFrom (when this version of the parcel took effect) and bbox. Set include_geometry for each boundary as GeoJSON (keep limit small). Use MN_parcel_image to see a parcel and its neighbours.
query*string
Parcel id (10 digits), 'latitude, longitude', or Cyrillic address words
au2string
District code or prefix, to narrow an address search (04 = Ulaanbaatar)
radiusnumber
Search radius in metres around coordinates(default 25)
limitinteger
Maximum results(default 20)
offsetinteger
Skip this many address matches, to page through them(default 0)
include_geometryboolean
Include each boundary as a GeoJSON MultiPolygon(default false)
MN_parcel_image
A map image centred on a Mongolian parcel (id, the 10-digit parcel id from MN_parcel_search) or a point (query), with every parcel in view outlined and labelled with its parcel id, so you can see which parcel is where. The parcel passed as id is drawn in red with a larger label; a query point is a red dot. Returns the parcels in view as JSON (label, id, landUse in Mongolian, area in m²; labelled: false where its label didn't fit), then one JPEG per basemap. Basemaps: satellite (Esri World Imagery, often a few years old), hybrid (the same imagery with road and place names), streets and topo (Esri's street and topographic maps). Pass several to compare, e.g. ['satellite', 'streets'] to match buildings with street names. Parcel outlines are yellow on imagery and magenta on maps. Imagery is not always aligned with the cadastre: offsets of a few metres are possible. A scale bar is at bottom left. Labels are left off where they'd overlap; call again with a smaller radius to read them. Each image costs about as much context as 1,000 words, so ask for what you need.
idstring
Parcel id (10 digits), drawn in red
querystring
'latitude, longitude' to centre on instead of a parcel
radiusnumber
Metres from the centre to the image edge. Default: fits the parcel with a margin, or 80 m around a point
basemapsarray
One image per basemap(default ["satellite"])
sizeinteger
Image width and height in pixels(default 768)

Try asking

About the data