Multi-Agent System for Industrial Automation

Using Model Context Protocol (MCP) servers to allow autonomous agents to interact with a manufacturing digital twin, querying real-time sensor data.

AutomationAgentsMCPDigital Twin

A demonstration of how autonomous AI agents can safely interact with a manufacturing digital twin. By exposing the digital twin (Knowledge Graph) through a governed MCP server, agents can query real-time sensor data and trigger maintenance workflows while remaining strictly within their ontological constraints.

Work in progress — this project is currently in the conceptual design phase.