Zenity, the leading end-to-end security and governance platform for AI agents, announced an integration with Claude’s Compliance API that extends governance and security controls for organizations using Claude Enterprise, Anthropic’s enterprise AI platform.
As AI agents become increasingly embedded in enterprise operations, security teams face growing challenges around visibility, oversight and risk management. Across Claude Enterprise, AI agents are taking on increasingly operational roles. Claude Code generates and commits production code. Cowork acts on behalf of employees across connected business applications. Claude Chat supports decisions across workflows that touch sensitive data and systems. This shift is creating new governance and security requirements for enterprises.
“Security teams are quickly realizing that the challenge is no longer what an AI model says, but what an AI agent can do,” said Ben Kliger, CEO and co-founder of Zenity. “Agents can access systems, invoke tools, interact with sensitive data and take actions across enterprise workflows. As they become embedded in day-to-day operations, organizations need visibility into those actions, governance over the tools agents use and security controls that help prevent risky behavior before it impacts the business.”
Also Read: 1Kosmos Delivers Workforce Identity Verification and Passwordless Authentication Through Google Cloud Marketplace
Using Zenity with Claude Enterprise, organizations gain visibility into agent activity, tool invocations and configuration settings across their Claude environments. The platform provides full-lifecycle coverage for Claude Code, Cowork and Chat, where agents can invoke tools, access enterprise systems and act on behalf of users. Security teams can discover and govern Model Context Protocol (MCP) servers, plugins and skills, maintain detailed audit trails of agent actions and monitor how agents interact with enterprise systems and data.
The integration also helps organizations detect AI-specific threats, including prompt injection attempts, credential exposure and unauthorized agent actions. This enables security teams to govern not only what AI agents say, but also the actions they take across enterprise workflows.
Key benefits include:
- Visibility into agent activity, tool invocations and configurations across Claude Enterprise
- Governance of Model Context Protocol (MCP) servers, plugins and skills
- Detection of prompt injection attempts, credential exposure and unauthorized agent actions
- Detailed audit trails of agent activity for compliance, investigations and risk management
- Security controls designed to help prevent risky agent actions across enterprise systems, data and tools
SOURCE: Businesswire