Netskope Extends Security Controls to AI Agent Traffic

Netskope is addressing that gap by extending the Netskope One platform to secure MCP communications across the enterprise. The update focuses on visibility, access control, and risk assessment, giving security teams a way to see which MCP servers and clients are in use and how agents are interacting with enterprise systems. With thousands of public MCP servers now available, this kind of baseline visibility is quickly becoming table stakes.

As enterprises move beyond copilots and into agentic AI, new plumbing is forming behind the scenes. Model Context Protocol, or MCP, is quickly becoming the standard way AI agents connect to enterprise data and tools. That shift brings real operational value, but it also introduces a new class of security exposure. MCP traffic can execute actions, move data, and operate autonomously, often outside the visibility of traditional controls.

At a policy level, the platform brings MCP traffic into familiar security workflows. Organizations can apply least privileged access, block MCP connections by default where needed, and monitor non human traffic moving between agents, tools, and data sources. MCP activity is logged at a granular level, from session initiation to tool requests and responses, with sensitive data inspection extending to MCP-driven workflows.

The bigger takeaway is that MCP is no longer theoretical. As agents move from experimentation to production, security teams need guardrails that understand how agents actually operate. By treating MCP as a first-class security surface, Netskope is positioning itself early in what will likely become a standard requirement for enterprises building and deploying AI agents at scale.