Netskope introduced new security capabilities that give enterprises visibility and control over Model Context Protocol (MCP), the fast-growing standard that links AI agents to corporate data and internal tools. MCP exposes powerful automation pathways, but also introduces new risks as thousands of publicly accessible MCP servers appear across the internet. Netskope’s enhancements extend its Netskope One platform to monitor MCP traffic, enforce least-privilege access, and identify sensitive data moving through AI agent workflows.
The new capabilities allow security teams to detect MCP servers and clients in real time, evaluate their risk profiles, and block or permit AI agent interactions using granular context-based policies. Enterprises can monitor non-human MCP traffic across agents, tools, hosts, and data sources, and log detailed MCP events including tool requests, responses, initializations, and deployments. Netskope also applies its Cloud Confidence Index (CCI) to MCP resources, providing a risk score to prioritize which AI tools and agent integrations require closer scrutiny.
By integrating these controls into Netskope One, organizations can prevent data leakage, enforce compliance, and prepare governance frameworks for expanding agentic AI use cases. The MCP security capabilities are now in Preview and are expected to reach general availability in the first half of 2026. Netskope is demonstrating the functionality at AWS re:Invent, where AI governance and agent-based automation remain top enterprise priorities.
• Real-time identification of MCP servers and clients, including attributes such as version, host, protocol, and data source
• Extension of Netskope’s CCI risk scoring to MCP servers
• Granular access controls with default block options for MCP traffic
• Detection and monitoring of non-human traffic across agents, tools, and data pipelines
• Logging of MCP sessions, initializations, tool requests/responses, and deployments
• Identification of sensitive data—including IP and credentials—in MCP interactions
“Every team wants to confidently accelerate AI adoption, and emerging protocols such as MCP are now fundamental to that discussion,” said John Martin, Chief Product Officer, Netskope.
🌐 Analysis
Security vendors are rapidly adapting to the rise of agentic AI frameworks, and MCP has become an early focal point as enterprises begin wiring LLM-based agents directly into business systems. Netskope’s move aligns with increased investment in AI-aware security controls across the industry, particularly as CSPM, SASE, and data-loss prevention platforms incorporate agent traffic inspection. Competitors are expected to follow with their own MCP-aware policies as enterprise AI pilots evolve into production deployments in 2026.
