AI-Driven DevSecOps: Securing Coding Agents with Model Context Protocol (MCP)

As DevOps AI tooling evolves, DevSecOps and cloud-native security controls are reaching more teams, faster than ever. Product and security teams are no longer constrained by their understanding of tools, command-line interfaces, integrations, and policy languages. With the right AI-driven DevSecOps toolchain, we are able to install, configure, and execute DevSecOps tools and policy controls using prompts.

Join SANS Certified Instructor Jon Zeolla for Part 1 of the AI-Driven DevSecOps series. In this webcast, Jon will cover AI coding agents, their baseline configuration, and how extensions such as Model Context Protocol (MCP) extend their functionality. Live demonstrations will show you how to configure an OpenVSCode server's Continue AI extension to use an OpenAI-compatible Large Language Model (LLM). Then, extend the model's capabilities using MCP servers with access to different tools, command-line interfaces, and APIs.

This webcast supports content and knowledge from SEC540: Cloud Native Security and DevSecOps Automation. To learn more about this course and explore upcoming sessions, Click Here.

Watch the rest of this series:

• AI Driven DevSecOps Part 2: Protecting Kubernetes Microservices with the Kong Ingress Controller
• AI Driven DevSecOps Part 3: Agentic Code Review in GitLab & GitHub
• AI Driven DevSecOps Part 4: Observing Microservices with OpenTelemetry and Grafana
• AI Driven DevSecOps Part 5: Zero-Downtime Deployments with Kubernetes