Tag
AgentAudit is an open, extensible framework for evaluating the full lifecycle of AI agents across capability, grounding, security, and behavioral dimensions, enabling precise failure attribution and highlighting trustworthiness differences among various language models.