PRE Security Announces AgentGuard for AI Agent Security
The platform capability assesses agent behavior and can block actions or require human approval before execution.
- By Chelsey Arries
- Oct 08, 2026
PRE Security has announced AgentGuard, a new capability within its security platform designed to monitor autonomous artificial intelligence agents, identify departures from assigned tasks and intervene before actions are executed.
Part of the PRE Security Platform, AgentGuard evaluates an agent’s identity, purpose and activity history. The company said it analyzes sequences of behavior to assess potential next steps, rather than reviewing individual prompts alone.
Positioned between agents and their tools, the system uses behavioral analysis and threat intelligence to determine whether actions should proceed, receive monitoring, need human approval or be blocked.
PRE Security said a demonstration involving an accounts-payable agent stopped an attempted external data transfer. In a separate coding-assistant demonstration, a command permitted during routine activity was restricted after suspicious behavior.
The company is also integrating AgentGuard with other investigation and monitoring capabilities on its platform to connect intervention with follow-up analysis and monitoring.
About the Author
Chelsey Arries is assistant editor of Security Today, Campus Security Today, Occupational Health & Safety and Environmental Protection Online.