# Brex developed a new approach to AI agent security Creating security policies by observing agent behaviors. - **Source:** [VentureBeat](https://venturebeat.com) - **Published:** 2026-07-17 - **Topic:** [Data & AI](https://brifing.si/c/data-ai) - **Original article:** - **On Brifing:** https://brifing.si/s/e30f868a-e0dd-4434-a147-e3ace3249131 ## Introduction Brex realized traditional methods were insufficient for securing AI agents. They developed a new approach by observing real agent behaviors, aiming to minimize security vulnerabilities. ## Limitations of Traditional Methods Traditional security measures like API permissions and model constraints limited agent flexibility. Brex found these methods inadequate for managing real-world agent behaviors. ## CrabTrap Platform Brex developed an internal platform called CrabTrap. It intercepts all network traffic, reviews policy rules, and uses an LLM judge to approve or deny requests. ## Importance of the Transport Layer Brex identified the transport layer as an overlooked opportunity for security. CrabTrap operates at this layer, offering a language and API-agnostic solution without requiring integrations. ## LLM Judge System CrabTrap combines static rules with an LLM judge. The judge only activates for unknown endpoints or unusual requests, preserving performance. ## Policy Creation Process Brex creates policies based on observed behaviors rather than writing them from scratch. This approach ensures policies are more effective and realistic. ## Evaluation System Policy changes undergo an evaluation system before going live. This system compares historical audit logs to report changes. ## Live Feedback Loop Brex developed a feedback loop for continuously denied resources. It notifies humans or agents to propose policy updates. ## Conclusion and Significance Brex's approach sets a new standard in AI agent security. Behavior-based policies enhance security while maintaining agent functionality. ## Sources covering this story - https://venturebeat.com/orchestration/brex-built-its-ai-agent-policy-by-watching-what-agents-actually-do-not-by-writing-rules-first — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._