This research introduces a novel approach for creating autonomous security agents with configurable personality traits that influence their decision-making and behavior.

• Builds on SANDMAN architecture that leverages the Five-Factor OCEAN personality model
• Demonstrates how personality induction significantly impacts agent task planning
• Enables more predictable and customizable AI agent behaviors for security applications
• Advances proactive cyber defense strategies through personality-driven agents

This research matters because it allows security teams to develop more sophisticated autonomous defense systems with predictable behavioral patterns tailored to specific security contexts.

Original Paper: Personality-Driven Decision-Making in LLM-Based Autonomous Agents