Knowledge of AI and Generative AI security, including LLMs, RAG, vector databases, embeddings, AI agents, tool use, AI-assisted code generation, prompt injection, data leakage, model and RAG poisoning, agent privilege escalation, insecure tool integration, AI supply-chain risk, and security controls for prompts, context, embeddings, memory, outputs, knowledge sources, models, plugins, tools, and supporting infrastructure.