Formal Runtime Verification for Tool-Using LLM Agents: An Offline Same-Benchmark Study on AgentDojo and STAC

Nikolaos Kekatos, Stylianos Basagiannis, Marinelio Chintri +5

Guardrails for tool-using LLM agents are usually application-specific rules, which makes multi-step, data-dependent safety policies hard to specify, audit and reuse. As a declarati…