Runnable multi agent case study with orchestrator to worker routing, selection disambiguation, agent registry, and audit log

Semantic testing for Microsoft Copilot Studio agents using Pytest and DeepEval G-Eval metrics (LLM-as-a-Judge). Generates interactive HTML reports for agent response quality.

End-to-end Copilot Studio chatbot powered by FastAPI, Azure AI, and cloud services with RAG and Docker/Jenkins deployment.

🤖 Demonstrate building and governing safe, scalable multi-agent systems with the Iron Legion framework for clear, repeatable business use.