MLIT DPF MCP

Overview

MLIT DPF MCP is the official Model Context Protocol (MCP) server (α version) developed by the MLIT DATA PLATFORM team under Japan's Ministry of Land, Infrastructure, Transport and Tourism (国土交通省). It wraps the public APIs of the 国土交通データプラットフォーム — a centralized hub that aggregates MLIT-held data with private-sector datasets for unified search, visualization, and download.

By running this MCP server, large language models (such as those in Claude Desktop) can interact directly with the platform in conversational Japanese or English. Users can issue vague or complex natural-language instructions (e.g., "Find recent land price data near Tokyo stations" or "Retrieve disaster prevention datasets for coastal areas") and receive structured results, metadata, and download links — all without needing deep API knowledge.

Key Features

• Natural Language Data Access: Translate ambiguous user queries into precise API calls against the MLIT Data Platform.
• Tool Exposure: Provides multiple MCP tools for dataset discovery, filtering, metadata retrieval, and content download.
• Seamless MCP Integration: Works with any MCP-compatible host/client (Claude Desktop and similar).
• Python-based Server: Easy local deployment with Python 3.10+.
• Government Data Focus: Covers infrastructure, geospatial (PLATEAU-compatible), land use, transportation, disaster prevention, real estate, and more.
• Secure & Official: Backed by MLIT; requires a free MLIT Data Platform API key for authenticated access.

Requirements

• OS: Windows 10/11 or macOS 13+
• Python 3.10+
• MCP Host: Claude Desktop or other compatible clients
• Recommended: 8GB+ RAM
• MLIT API Key (obtained via free account on data-platform.mlit.go.jp)

Installation & Setup

1. Clone the repository:
   git clone https://github.com/MLIT-DATA-PLATFORM/mlit-dpf-mcp.git
   cd mlit-dpf-mcp
   

2. Create and activate virtual environment.
3. Install dependencies.
4. Configure your MLIT_API_KEY and optionally MLIT_BASE_URL.
5. Add the server configuration to your MCP host (e.g., Claude Desktop's mcpServers section pointing to src/server.py).

Detailed Japanese/English instructions and configuration examples are available in the repository README.

Use Cases

• Urban Planning & Research: Quickly gather land price, building, or transportation data for analysis.
• Disaster Preparedness: Retrieve hazard maps and prevention datasets via simple prompts.
• Real Estate Insights: Query geospatial and property-related information conversationally.
• Government Data Exploration: Enable non-technical users and AI agents to explore Japan's public infrastructure datasets.
• Multi-Agent Workflows: Combine with other MCP servers (e.g., PLATEAU data) for richer geospatial AI applications.

Technical Details

• Protocol: Model Context Protocol (MCP) — standardized tool, resource, and prompt exposure.
• Language: Python.
• License: MIT (as per PyPI/package metadata).
• Status: Alpha version; actively maintained by the official MLIT team.

This MCP server exemplifies how government open data initiatives can leverage emerging AI protocols to democratize access to complex datasets.

For the latest setup guide, examples, and updates, visit the GitHub repository.