Japan Geospatial Open Data Mcp

io.github.atushi1841Published by GitHub account atushi1841v1.0.6Updated Oct 24, 2026

Japan geospatial MCP: Apify:apify.com/fruitful_quintessence | Gumroad:atushi5.gumroad.com/l/kutuxe

VerifiedSTDIODesktop onlyLocation & LifestyleData & Analytics

Overview

AI-generated overview

A local MCP server that gives an AI assistant Japanese national-scale geospatial open data: address geocoding, GSI map tiles, e-Stat statistics and school-distr

What you can do
Geocode Japanese addresses to latitude and longitude via the GSI address search API, and fetch GSI XYZ map tiles (standard, pale, blank, aerial, vector) as base64 images at a given zoom and location. List all accessible data sources and their key requirements. Retrieve e-Stat statistics such as population, population change and households, plus statistical table metadata. List nationwide elementary school district files (GML/GeoJSON) from MLIT land numerical information, and fetch school district GeoJSON tiles from the MLIT Real Estate Information Library.
When to use it
Useful for Japanese real-estate analysis (school districts, zoning), education infrastructure assessment, disaster prevention and urban planning work where an assistant should pull official Japanese geospatial and statistical open data. Core geocoding, tiles and school district listing work without any keys.
Requirements
Runs as a local Python process installed from source (git clone, virtual environment, pip install). Optional environment variables: ESTAT_APP_ID for e-Stat statistics, REINFOLIB_API_KEY for the Real Estate Information Library, and USER_AGENT to override the default user agent. Network access to GSI, e-Stat and MLIT services is required.
Before you install
Two tools need registered credentials: ESTAT_APP_ID (free e-Stat application ID) and REINFOLIB_API_KEY (MLIT Real Estate Information Library key); store them in the client configuration. Requests go to third-party public services (GSI, e-Stat, MLIT). GSI tile terms require crediting the Geospatial Information Authority of Japan in output. The README also links to Apify and Gumroad storefronts, which are separate from the server itself.

Installation

In SourceWeft

  1. Open Japan Geospatial Open Data Mcp in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

japan-geospatial-open-data-mcp

日本全国規模のジオスパティアル・オープンデータを AI エージェントに届ける MCP Server。 GSI 地理院タイル・地理院住所検索・e-Stat 統計 API・国土数値情報（小学校区）・不動産情報ライブラリを 1 つの MCP でまとめて公開する。

Context: FOSS4G Hiroshima 2026 発表「Building an MCP Server Ecosystem on Japan's National-Scale Geospatial Open Data Platform」(Mami Enomoto, MLIT) が示した方向性に追従するエコシステム構築タスク。 Needs: 不動産分析（学区・用途地域）・教育インフラ評価・防災・都市計画。 TOS: GSI/e-Stat/国土数値情報は公式オープンデータ・再利用自由。

Tools (7)

ToolDescriptionKey required
geocode_japan住所 → 緯度経度（GSI 住所検索 API, msearch.gsi.go.jp）No
get_gsi_tile地理院 XYZ タイル（std/pale/blank/aerial/vector）を base64 画像で取得No
list_data_sourcesアクセス可能な全データソース一覧と鍵要件No
get_estat_datae-Stat 統計データ取得（人口・人口動態・世帯 etc.）ESTAT_APP_ID
get_estat_metae-Stat 統計表メタ情報取得ESTAT_APP_ID
list_school_districts全国小学校区 (A27) GML/GeoJSON ファイル一覧（国土数値情報, nlftp.mlit.go.jp）No
get_school_district_geojson小学校区 GIS タイル (XKT004) GeoJSON 取得（不動産情報ライブラリ）REINFOLIB_API_KEY

Environment variables (optional)

No keys → core GSI + school district listing still works out of the box.

Install

bash
git clone https://github.com/atushi1841/japan-geospatial-open-data-mcpcd japan-geospatial-open-data-mcppython3 -m venv .venv && source .venv/bin/activatepip install -r requirements.txt

Claude Desktop config

json
{  "mcpServers": {    "japan-geospatial-open-data": {      "command": "/path/to/japan-geospatial-open-data-mcp/.venv/bin/python",      "args": ["/path/to/japan-geospatial-open-data-mcp/src/server.py"],      "env": {        "ESTAT_APP_ID": "your-e-stat-app-id",        "REINFOLIB_API_KEY": "your-reinfolib-key"      }    }  }}

Example queries

  • 「東京都渋谷区の小学校区はどれか」 → geocode_japan("東京都渋谷区") → list_school_districts
  • 「渋谷区の航空写真 2024 年をズーム 15 で表示」 → get_gsi_tile(layer="nendophoto2024", lat=35.657, lon=139.698, z=15)
  • 「日本の 2020 年人口（年齢別）」 → get_estat_data(table_key="population")
  • 「地価公示ポイントの GeoJSON」 → (future: get_reinfolib_xkt004 系)

Data sources & licensing

SourceURLLicense
GSI tiles & geocoderhttps://maps.gsi.go.jp/GSI tile terms (free, credit required)
e-Stat APIhttps://www.e-stat.go.jp/api/Open statistics (free, key registration)
MLIT Kokuto Suuchi Jhouhouhttps://nlftp.mlit.go.jp/ksj/MLIT open data (free)
MLIT Real Estate Information Libraryhttps://www.reinfolib.mlit.go.jp/MLIT open data (free, application)

Credit requirement: when using GSI tiles, include "Geospatial Information Authority of Japan" in your output.

MCP Registry

Publishing to the official MCP Registry is a follow-up task. The mcp-publisher CLI (GitHub auth, io.github.atushi1841/japan-geospatial-open-data-mcp namespace) is the standard flow.

Roadmap / future tools

  • 用途地域 (zoning) GeoJSON from reinfolib
  • 洪水浸水想定区域 / 土砂災害警戒区域 hazard layers (reinfolib)
  • 地価公示 point API
  • e-Stat 250m mesh population for density analysis
  • 中学校区 (A28) listing
  • Bulk GeoJSON download helper for nlftp files (A27 prefecture-level GML)

Apify Store

This MCP server is published as an Apify Actor: japan-geospatial-open-data-mcp.

Apify Store: atushi1841

Data & Store

Source: README.md at commit e63caf6

Tools

0
SourceWeft started this package once in an isolated sandbox, without credentials, to list its tools. That failed: the package type cannot run in the sandbox.

Version history

4
  1. v1.0.6LatestOct 24, 2026
  2. v1.0.5Oct 23, 2026
  3. v1.0.4Oct 22, 2026
  4. v1.0.0Oct 17, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts