Netskope announced its upcoming integration with Amazon Bedrock AgentCore to extend security controls into agentic workflows. The integration aims to provide enterprise security teams with consistent allow-or-deny decisions on every agent's action by combining Netskope's detection capabilities with deterministic enforcement.

Amazon Bedrock AgentCore is a platform designed to build, connect, and optimize agents at scale. Its policy engine enforces real-time, deterministic controls at the gateway across all agent traffic, including agent-to-tool, agent-to-LLM, and agent-to-agent communications.

Through this integration, Netskope One AI Guardrails will gain the ability to provide complementary detection capabilities. These include prompt injection detection, sensitive data exposure protection, toxic output filtering, restricted topic enforcement, and model response validation. These signals feed into the AgentCore policy, which acts on them at the gateway, outside the agent's reasoning loop.

John Martin, Chief Product Officer at Netskope, emphasized the need for a new security model in the agentic era. He stated that the collaboration with AWS allows organizations to move AI agents into production with confidence, ensuring that what an AI agent is allowed to do aligns with what it actually does.

Netskope continues to collaborate with leading frontier AI models to secure organizations at AI speed. These efforts include Anthropic's Project Glasswing and OpenAI's Trusted Access for Cyber (TAC) program. The company is recognized as a trusted defender of critical, AI-ready enterprise infrastructure.

Security and networking practitioners can learn more about transitioning from AI experimentation to high-performance innovation by registering for Netskope's global roadshow series, AI in the Fast Lane.