Japan geodata MCP server for cadastral parcel lookup, reverse geocoding, and address matching.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent — or use 1-click editor setup below.
We ran the install command below but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.
npx -y geojp-mcpNo response to initialize.
This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 19d ago.
💡 Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Geojp.
get_parcelChiban (cadastral lot number), ōaza/aza, municipality, and survey precision class for the containing parcel
reverse_geocodeNearest residential address (prefecture, municipality, district, postal code)
geocodeUp to 10 candidate locations for a partial address match
lookup_locationAddress and chiban together in one call (an orchestration example — succeeds with address-only if the point falls outside ChibanJP's current coverage)
The geojp MCP server connects AI agents to two Japan-focused geodata services: ChibanJP for cadastral parcel information and ReverseGeoJP for address geocoding. Its tools cover both sides of a location workflow: coordinates can be converted into a residential address, a cadastral lot number, or both, while a partial address can be turned into candidate coordinates.
The distinction between residential addressing and chiban is important in Japan. A residential address is used for maps and mail, while chiban is the official cadastral lot number used in property registries, land transactions, and legal documents. These systems do not map predictably to one another, so an application that needs registry-oriented location data may need a dedicated parcel lookup.
Four MCP tools are defined in shared tool code and can be served through either of two entry points:
get_parcel sends latitude and longitude to ChibanJP and returns the containing parcel’s chiban, ōaza/aza, municipality, and survey precision class.reverse_geocode sends coordinates to ReverseGeoJP and returns the nearest residential address, including prefecture, municipality, district, and postal code.geocode accepts an address string and returns up to 10 candidate locations for a partial match.lookup_location calls both services in parallel and returns address and chiban data together. If the coordinate is outside ChibanJP’s current coverage, it can still succeed with address-only output.All four tools use one API key. The key is obtained from reversegeojp.com and is passed to the underlying services by the MCP server.
For local use, install the package globally with npm install -g geojp-mcp, or install dependencies from source and run the project build. Set GEOJP_API_KEY in the MCP client’s environment. In the stdio arrangement, the client launches npx geojp-mcp as a child process.
Claude Code can register the local process with claude mcp add, passing the key with -e GEOJP_API_KEY=.... The README also provides an .mcp.json or ~/.claude.json configuration block. Claude Desktop uses the same mcpServers entry in claude_desktop_config.json; restart the application after adding it.
The hosted option is available at https://mcp.reversegeojp.com. It uses streamable HTTP and expects the key in an Authorization: Bearer ... header, so no local installation or build is required. The /health endpoint provides a basic health check. Local Worker development and deployment are available through the project’s worker:dev, worker:check, and worker:deploy scripts.
The geojp MCP server is suited to workflows that need Japan-specific location enrichment, such as checking a parcel associated with coordinates, adding a postal address to a map point, matching an incomplete Japanese address, or retrieving both address and cadastral context in one call. lookup_location is an orchestration example rather than a separate geodata API; it combines the two backing services.
A smoke-test script can verify that the tool list loads and that API calls are wired correctly. It uses a dummy key and does not validate successful data access; its expected API response is an invalid_api_key error.
ChibanJP does not necessarily cover every coordinate. When a point is outside its current coverage, lookup_location may return only the residential address. Chiban and residential address values should therefore be treated as different data fields rather than interchangeable formats.
The remote Worker is stateless and does not retain session state. Each request reads its API key from the authorization header and proxies to the production APIs. A real API key is required for data calls; the smoke test is only a wiring check. The project is released under the MIT license.
Factual signals from GitHub, npm, and our automated checks — not a rating.
No reviews yet — be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/geojp)<a href="https://allmcps.com/mcp/geojp"><img src="https://allmcps.com/api/badge/geojp?style=directory" alt="Geojp on AllMCPs" /></a>