Zenity, which provides in-depth security and protection from existing and emerging threats from agents that are built across an organization, has expanded its relationship with Microsoft to offer integrated native, inline attack prevention for Microsoft Copilot Studio agents. Zenity is the first security and governance platform purpose-built for AI agents — spanning SaaS, home grown platforms (Cloud), and end-user devices (Endpoint). The integration is designed to help organizations deploy agentic AI at scale with greater control and security. Zenity can now operate within each agent built in Copilot Studio.
By combining Microsoft Copilot Studio’s flexibility with Zenity’s security-first approach now incorporated into inline behaviors and actions of agents, businesses can confidently and safely deploy AI agents as part of the integration. This provides inline controls on tool invocation, such as MCP servers, CRM systems, business applications, and email to prevent data exfiltration, direct and indirect prompt injections, and improper secrets handling. Organizations can also now apply native security controls as the backbone for expanding Copilot Studio across departments and line of business users.
“This integration marks a major step forward in enabling secure and scalable AI innovation for enterprises,” said Shay Gadidi, vice president of product management at Zenity. “By combining Microsoft Copilot Studio’s flexibility with Zenity’s security-first approach now incorporated into inline behaviors and actions of agents, businesses can confidently deploy AI agents without sacrificing control.”
Zenity helps block actions that show signs of potential security risks, both inline and in real time, by analyzing intent and behavior signals before they complete. Together, Microsoft and Zenity are working to provide customers full execution context for every agent, ensuring compliance and enabling automated enforcement of enterprise policies. The inline threat prevention capabilities also extend Zenity’s platform to deliver broader threat analysis, more advanced intelligent threat reasoning and robust threat disruption into AI agent activity.
“With this integration with Zenity, we’re making it easier for enterprises to confidently scale AI agent development across their organizations,” said Shay Gurman, vice president, Microsoft Copilot Studio. “Together, we are combining Microsoft Copilot Studio’s intuitive agent-building experience with secure inline risk prevention. In doing so, we are empowering more users, from professional developers to business users, to innovate safely and responsibly while maintaining the security and governance that enterprises expect.”
With this integration, customers can expect to benefit by scaling Copilot Studio across business units to build agents for teams while maintaining centralized security and governance. This minimizes the risk of breaches, data loss or policy violations without disrupting workflows. It also ensures proper configuration from the start, accelerating deployment and productivity while gaining deep insights into how agents are built and customized in Copilot Studio.
“But this democratization of AI development introduces new and critical risks without active administration of the system,” said Andrew Silberman, Zenity’s Director of Corporate Marketing, who is responsible for effectively communicating Zenity’s unique positioning and value to customers and partners.
“Anyone can build an agent, including those without security or technology expertise,” he said. “Anyone can easily configure agents to interact with a variety of tools – MCP servers, CRM systems, business applications, email, other agents, data sources, and lots more. These agents can be manipulated to act unpredictably and/or maliciously, particularly when configured to invoke other tools.”
This is why Zenity’s inline prevention capabilities are designed to disrupt and prevent risky agent behavior before damage occurs. Security teams are able to intercept agent tool invocation in real time, analyzing intent and behavior to prevent risk. They can disrupt risky behavior, stopping data exfiltration and unauthorized access before it happens. They can also trace root causes, identifying misconfigurations, privilege misuse, and insecure triggers.
For partners, this provides seamless integration with top Agentic AI and business-development platforms, access to the latest technological advancements, and enhanced security features for specific technologies.
In our latest demo which we demonstrated at BlackHat 2025, we showed how a misconfigured agent, even one built with good intentions, can become a serious security liability. Zenity’s integration with Copilot Studio represents a shift towards intent-aware, behavior-driven security – one that operates in-line, not after the fact.
”The companies winning with AI aren’t just deploying agents faster – they’re operationalizing them responsibly,” said Dina Durutlic, Principal Product Marketing Manager at Zenity. “They realize AI agents are creating a new, dynamic attack surface that traditional tools were never designed to handle. These agents span the entire enterprise ecosystem. For example, Microsoft 365 Copilot, Copilot Studio, and Salesforce Agentforce are SaaS‑managed agents. GitHub Copilot, Cursor, and Claude desktop run directly on user devices as device‑based agents.
“Zenity helps security teams get ahead of these risks,” Durutlic continued. “We surface where agents are running, what they have access to, and how they interact with users and data giving you the ability to govern usage, apply role-based guardrails, and prevent sprawl before it starts.
The integration is now available in public preview through the Zenity listing on Azure Marketplace.