Skip to content

MCP Security Summit Workshop

Launch MCP Security Workshop

Ready to put the OWASP MCP Top 10 security guidance into practice? The MCP Security Workshop provides hands-on labs and exercises that walk through implementing secure MCP servers on Azure.

MCP Security Workshop

The workshop is open source and designed for both individual learning and team training sessions.

Workshop Overview

The workshop covers:

  • Practical Implementation: Step-by-step guidance for building secure MCP servers using Azure services
  • Security Controls: Hands-on exercises implementing the security patterns described in this guide
  • Real-World Scenarios: Labs based on common enterprise deployment patterns
  • Azure Integration: Deep-dive on Azure services like Managed Identity, Key Vault, API Management, and more

What You'll Learn

The workshop reinforces the security concepts from this guide through practical implementation:

  • Setting up secure authentication with Managed Identity
  • Implementing proper secret management with Key Vault
  • Configuring network isolation with VNets and Private Endpoints
  • Adding telemetry and monitoring with Application Insights
  • Implementing authorization and access controls
  • Testing for MCP-specific security threats

Prerequisites

To complete the workshop exercises, you'll need:

  • An Azure subscription
  • Basic familiarity with Azure concepts
  • Understanding of the OWASP MCP Top 10 (covered in this guide)

Getting Started

  1. Read the Guide: Review the OWASP MCP Top 10 sections relevant to the labs you want to complete
  2. Access the Workshop: Visit azure-samples.github.io/sherpa
  3. Follow the Labs: Complete the hands-on exercises at your own pace
  4. Apply to Production: Use the patterns and code from the workshop as templates for your own MCP deployments

Feedback and Contributions

The workshop is continuously evolving based on community feedback and emerging patterns. If you have suggestions or want to contribute new labs:

  • Workshop Repository: Contributions welcome on GitHub
  • Issues and Discussions: Share your experience and help improve the content

Next Steps: