"Netskope introduces security capabilities for Model Context Protocol (MCP) communication. The functionality is designed to help organizations deploy AI agents securely. MCP communication enables AI systems to connect to business data and issue autonomous commands. MCP was launched by Anthropic in November 2024 as an open standard for connecting AI agents to external systems. The protocol acts as a universal adapter between AI applications and business assets."
"But that growth also comes with risks. Because MCP communication can give AI systems access to sensitive business data and enable autonomous commands, new security challenges arise. This is a particular concern with thousands of publicly available MCP servers. The new capabilities in the Netskope One platform provide organizations with complete visibility into MCP tool usage. The platform constantly identifies MCP servers and clients within the organization in real time, including attributes such as name, ID, URL, version, host, data source, and protocol."
MCP enables AI agents to connect to business data and execute autonomous commands as a universal adapter between AI applications and business assets. Adoption by major cloud providers and thousands of public MCP servers has expanded usage and exposure. That growth creates risk because MCP connections can expose sensitive business data and permit autonomous actions. Netskope One provides continuous discovery and identification of MCP servers and clients in real time, logging sessions, initializations, tool requests and responses, and non-human traffic. Netskope extends its Cloud Confidence Index to MCP servers, offers granular, context-based policy controls, a default MCP block option, and real-time data loss prevention.
Read at Techzine Global
Unable to calculate read time
Collection
[
|
...
]