Help Net Security reports that NinjaOne has launched Katana AI, an agentic AI technician built into its platform, alongside the NinjaOne MCP server. Katana AI draws on device telemetry, alert history, and existing resolution knowledge to surface likely root causes and recommend next steps, including remediation scripts, where technicians already work. The MCP server extends that capability to external AI agents, letting them interact with NinjaOne data and take action directly in the platform.
From a security and privacy standpoint, the notable design choice is governance. According to the announcement, external agents work through NinjaOne rather than around it, and are subject to the same role-based permissions IT teams already manage. The company says this gives organizations a single orchestration point and a single governance model for AI agents, and that customers can set what the Katana agent can access and approve its actions.
NinjaOne also emphasizes token efficiency for predictable costs, and cites other AI features such as vulnerability management and AI-summarized remote session recordings. The source is a single vendor announcement, so there is no independent assessment of these claims; the article presents them as product capabilities rather than verified outcomes.