Prompt Security capabilities will be integrated with AgentCore, providing runtime protections to enterprises deploying AI agents at scale
SentinelOne (NYSE: S), a global leader in AI-powered cybersecurity, announced its new integration with Amazon Bedrock AgentCore, which will be made available through SentinelOne's Prompt Security offering. This integration reflects SentinelOne's commitment to strengthening the security controls customers use in agent-based workflows, enabling companies deploying AI on AWS to secure and extend their AI-based workflows.
Amazon Bedrock AgentCore is the platform that helps you build, connect, and optimize agents at scale. Through its policy engine, AgentCore applies real-time deterministic controls at the gateway across all agent traffic, including agent-to-tool, agent-to-LLM, and agent-to-agent communications. With this integration, SentinelOne's Prompt Security will provide security insights covering prompt injection detection, exposure of personally identifiable information (PII), validation of tool use, monitoring of LLM responses, and defense against data leakage, which are passed to AgentCore's policy management system. AgentCore acts on these recommendations at the gateway level, outside of the agent processing cycle, thus enabling enterprise teams to apply SentinelOne security recommendations along with deterministic and automated control measures consistently across all agent activities. Because detection can be probabilistic, but the application remains deterministic, customers get consistent allow or deny decisions on every agent action, allowing organizations to adopt enterprise-scale AI agents with the confidence that existing SentinelOne security controls apply automatically in every agent workflow.
“Agentic AI has become central to how businesses operate,” said Melissa K. Smith, SVP of Global Strategic Partnerships and Initiatives at SentinelOne. "Every organization that develops and manages AI agents on AWS needs the same security controls they leverage across the rest of their infrastructure. This upcoming integration applies those safeguards at the AgentCore gateway, where all agent traffic is evaluated, giving every organization the confidence to deploy AI agents knowing that existing SentinelOne policies apply automatically, from day one."
The new integration will be available on Amazon Bedrock AgentCore later this year. Customers with existing Prompt Security licenses will be able to apply them through the Bring Your Own License program. As enterprise AI agent deployments grow, so does consumption.
