Netskope Unveils New Security Features for Model Context Protocol to Ensure Safe AI Adoption in Organizations

Introduction

On December 1, Netskope (NASDAQ: NTSK), a leader in security and networking technologies for the cloud and AI era, has announced groundbreaking security features for the Model Context Protocol (MCP) communications. These newly introduced features are part of the Netskope One platform, designed to facilitate a secure environment for organizations looking to implement AI agents without compromising on security.

The Rise of MCP

As the adoption of AI continues to expand, MCP is swiftly emerging as the primary means for connecting AI agents to corporate resources. However, this transition brings about new security risks, as MCP communications allow AI systems to access corporate data and execute autonomous commands. With thousands of MCP servers publicly available, the urgency of addressing these risks has never been greater.

Key Features Unveiled

The enhancements to the Netskope One platform aim to safeguard communications with MCP-enabled AI, providing organizations with complete visibility into their MCP tool usage, managing access with minimum necessary permissions, protecting sensitive data, and ensuring compliance. Organizations using Netskope One will benefit from the following capabilities:

1. Real-Time Identification: Identify the MCP servers and clients in use within the organization in real-time, along with attributes such as name, ID, URL, version, host, data source, and protocol.

2. Risk Scoring with CCI: Apply the Netskope Cloud Confidence Index (CCI) risk scoring to MCP servers to quickly identify and prioritize AI tools, agents, and integrations that pose security or compliance risks.

3. Policy Controls: Implement context-based detailed policy controls, including default block options for MCP traffic, to manage access and prevent data leaks in real-time.

4. Traffic Monitoring: Detect and monitor non-human traffic (such as bots) between MCP servers, clients, tools, hosts, data sources, and development tools.

5. Event Logging: Keep track of MCP event logs for sessions, initializations, tool requests/responses, and deployments.

6. Data Identification: Identify sensitive data such as intellectual property and passwords being handled by MCP tools.

Expert Insights

John Martin, Chief Product Officer at Netskope, shared insights into the motivation behind these enhancements: “Every organization seeks to accelerate AI adoption in a secure environment. New protocols like MCP are central to this discussion. However, they also introduce security risks that traditional tools cannot handle. Therefore, we have expanded the leading features of Netskope One, enabling visibility, policy formulation, and risk assessment of MCP traffic to occur instantly. This is essential for organizations to utilize AI safely while developing agents for enhanced productivity.”

Availability

The new security features for MCP are currently available as a preview for Netskope customers and are set to be generally available in the first half of 2026. For more information on MCP security and updates from Netskope, readers are encouraged to visit the Netskope blog.

About Netskope

Netskope (NASDAQ: NTSK) leads the industry in modern security and networking technologies for the cloud and AI era. It delivers optimized access and real-time context-based security, catering to the needs of both security and network teams, wherever people, devices, and data reside. Trusted by thousands, including over 30 Fortune 100 companies, they rely on the Netskope One platform, Zero Trust engine, and robust NewEdge network to visualize usage across all clouds, AI, SaaS, web, and private applications, reducing security risks and enhancing network performance. For more details, visit netskope.com/jp.