Formally Specifying the High-Level Behavior of LLM-Based Agents

Open in new window