Continue the conversation

Got questions after watching? Join our Agentic AI Slack Community — where peers, experts, and innovators come together to tackle the toughest challenges in securing AI agents. Ask, share, learn, and help shape the future of agentic identity. 👉 Join the community today

Introduction

In this second chapter of the Agentic AI webinar series, Strata’s Chief Architect Granville Schmidt joins Mark Callahan to explain the Model Context Protocol (MCP)—the AI-native protocol powering agent-to-API communication.

You’ll get a deep dive into how MCP is being used by platforms like Claude and ChatGPT to access live enterprise data, and what this means for security, delegation, and identity management.

From delegated authorization to human-in-the-loop enforcement and real-time policy decisions, discover how Maverics Orchestration can secure and govern MCP workflows without rewriting your APIs.

What you’ll learn

  • What MCP is, how it started, and why it matters now
  • How AI agents use MCP to communicate with APIs and systems
  • Why OAuth alone isn’t enough to secure agent workflows
  • Where identity orchestration fits in securing and scaling MCP
  • How Strata’s MCP proxy and MCP bridge simplify adoption

Key Takeaways

  • MCP is the emerging standard enabling real-time data access for AI agents.
  • AI agents require delegated, traceable identity—not shared credentials.
  • MCP introduces new security gaps that traditional IAM can’t solve alone.
  • Maverics can act as an MCP proxy to offload authN, authZ, and audit.
  • Strata’s identity orchestration brings Zero Trust and human-in-the-loop enforcement to agent workflows.