KnowBe4 extends Agent Risk Manager to support Anthropic's Claude

KnowBe4 has extended its Agent Risk Manager to support Anthropic's Claude, expanding the security platform's AI agent governance layer beyond its existing native support for Microsoft Copilot.

Agent Risk Manager sits as a governance layer that monitors and governs agent execution without modifying the underlying model. KnowBe4 said the approach applies the same risk-based methodology it has used for human behaviour security for 15 years to what it calls the "agentic workforce".

With native support for both Claude and Copilot, the platform offers one-time setup for visibility across Claude.ai and API deployments; six specialised detection engines covering prompt injections, sensitive data and PII leaks, privilege escalations and unapproved tool access; and a blast radius visualisation tool - essentially an interactive network map that lets security operations teams identify high-risk nodes if an agent is compromised.

"It is incredibly important to protect what these agents are doing, especially when employees might be using features like skip permissions mode, which allows the LLM to make its own decisions autonomously with zero oversight," said Matt Duren, SVP of AI and data at KnowBe4.

"With KnowBe4's Agent Risk Manager for Claude, we focus on the actions, outputs, and tool usage of these agents, ensuring they do not become an unmonitored backdoor into sensitive company assets."