Multi-agent conversational (MAC) framework instantiating in-situ behavioral evaluations for LLM fairness - View it on GitHub