Amazon Bedrock AgentCore now supports Bedrock Guardrails in policy

Update existing AgentCore gateway deployments to incorporate Bedrock Guardrails, ensuring real-time evaluation of agent actions and inputs to detect and block security threats without requiring new infrastructure.

Adopt Bedrock Guardrails within AgentCore policy to provide deeper safety and security controls, enabling consistent enforcement and logging of policy evaluations for optimization and auditing.

Evaluate the integration of Bedrock Guardrails to enhance security and safety for AI agent workloads, reducing risks of prompt injection attacks and sensitive data exposure, and ensuring consistent enforcement across all agent actions.