What happened
Netskope's June 2026 release added MCP traffic parsing to the AI Gateway, enabling it to act as an inline intermediary between AI agents (MCP clients) and external tools/data sources. Agent Guardrails provide intent-, risk-, and access-based policy enforcement on agent traffic in real time.
Why it matters
Extends Netskope's inline SASE inspection to the MCP protocol layer — bringing the same DLP/threat-protection controls enterprises already use for web/cloud traffic to bear on AI agent communications, without requiring a separate point product.
Applicability
Existing Netskope customers deploying MCP-connected AI agents; evaluate as part of current renewal/upgrade cycle.