Fujitsu Social Digital Twin MCP Server

What is Fujitsu Social Digital Twin MCP Server?

This server integrates Fujitsu's Social Digital Twin and Digital Rehearsal API with the Model Context Protocol (MCP), enabling Large Language Models (LLMs) to access and simulate social and economic activities using natural language. It bridges the gap between LLMs and Fujitsu’s Digital Rehearsal simulation capabilities.

How to use Fujitsu Social Digital Twin MCP Server?

Install via Smithery or clone the repository and set up a Python 3.13+ environment with uv. Set two environment variables: FUJITSU_API_BASE_URL and FUJITSU_API_KEY. Start the MCP server with python -m fujitsu_sdt_mcp. For MCP-compatible clients like Claude Desktop, add the server configuration with the same environment variables. An interactive client (python client.py) is also provided for direct tool calls.

Key features of Fujitsu Social Digital Twin MCP Server

• Retrieve and display simulation lists
• Start simulations from natural language descriptions
• Retrieve and analyze simulation results
• Compare multiple simulation scenarios
• Analyze traffic simulations with region/time parameters
• Generate simulation configurations from natural language

Use cases of Fujitsu Social Digital Twin MCP Server

• Run a traffic optimization simulation for Tokyo’s morning rush hour and get CO2 emissions, travel time, and traffic volume.
• Compare different urban planning scenarios (e.g., signal timing changes vs. public transport frequency) before implementation.
• Generate simulation settings from a brief description, then start and analyze the simulation in one conversational workflow.
• Explore available simulation data and metrics interactively through an LLM chat interface.

FAQ from Fujitsu Social Digital Twin MCP Server

What is Digital Rehearsal?

Digital Rehearsal allows users to simulate human and social behavior in a digital space before implementing measures in the real world, enabling advance verification of effects and impacts.

What are the prerequisites to run this server?

Python 3.13 or higher, access to the Fujitsu API Gateway (API key), and an MCP-compatible LLM client (e.g., Claude Desktop).

How do I set up the required API credentials?

Set the environment variables FUJITSU_API_BASE_URL (default: https://apigateway.research.global.fujitsu.com/sdtp) and FUJITSU_API_KEY (your API key). You can also place them in a .env file.

What transport does the server use?

The server communicates with MCP clients using standard I/O.

What tools are available?

Tools include list_simulations, start_simulation, get_simulation_result, get_metrics, list_simdata, get_simdata, analyze_traffic_simulation, compare_scenarios, and create_natural_language_simulation_config.