security benchmark

llm agent (3) attack evaluation (1) knowledge manipulation (1) vulnerability analysis (1) adversarial evaluation (1) tool integration (1) large language model (1) model context protocol (1) prompt injection attack (1) vulnerability evaluation (1)