Security teams can now discover, observe, and secure every Copilot Studio agent, including shadow agents, from a single control plane.
SAN FRANCISCO, June 29, 2026 /PRNewswire/ -- Trust3 AI today announced that its Agent Control Plane, the operational core of the company's Agent DOS (Discovery, Observability, Security) framework, now integrates with Microsoft Copilot Studio. The integration gives security and AI platform teams a single place to discover every Copilot Studio agent, understand what each one is doing, and enforce security guardrails in real time.
Copilot Studio has made it easy for organizations to build and deploy AI agents. As adoption accelerates, security teams need visibility into every agent, the data it accesses, the tools it invokes, and the actions it performs. Trust3 AI addresses this challenge by continuously discovering Copilot Studio agents and governing their behavior without sitting in the data path.
With the integration, Trust3 AI extends its Agent Control Plane to Copilot Studio with:
Continuous discovery. Automatically discovers every Copilot Studio agent, including shadow agents built outside official inventories, along with ownership, connected data sources, and risk classification.
Tamper-evident observability. Captures prompts, tool calls, decisions, and execution history to provide complete replay and forensic investigation capabilities.
Runtime guardrails and kill switch. Enforces security policies during agent execution and enables security teams to immediately stop agent activity during an incident.
MCP content firewall. Protects Copilot Studio agents interacting with MCP servers by treating every MCP server as untrusted by default, limiting credentials per request, and removing prompt injection attempts embedded in tool descriptions or responses.
Identity-aware governance. Preserves the originating user identity throughout agent delegation so access controls, policy enforcement, and audit trails reflect the individual who initiated the request.
"Copilot Studio makes it easy to build AI agents. It should be just as easy to discover them, understand what they're doing, and stop them when they violate policy. That's exactly what our Agent Control Plane delivers."
Don Bosco Durai
Co-Founder and Chief Technology Officer, Trust3 AI
Trust3 AI will be demonstrating the Microsoft Copilot Studio integration during AI Engineer World's Fair 2026, taking place June 29 through July 2 in San Francisco.
The Copilot Studio integration is available now.
About Trust3 AI
Trust3 AI is the control plane for enterprise agentic AI, giving security and governance teams discovery, observability, and security across AI agents, the MCP servers they use, and the data they access. Built on the data governance foundation formerly known as Privacera, Trust3 AI provides consistent policy enforcement across Snowflake, Databricks, BigQuery, Microsoft, and more than 50 enterprise data platforms.
Learn more at https://trust3.ai
Recommended For You:
Are AI Agents for Insurance Safe to Use? Key Benefits and Risks to Consider
