Adrian: Runtime Security for AI Agents
How Adrian captures agent actions and reasoning, classifies them with a local LLM, and blocks or holds risky t
How Adrian captures agent actions and reasoning, classifies them with a local LLM, and blocks or holds risky t