GitHub Copilot has moved beyond simple code completion into agentic workflows across the end-to-end developer lifecycle. In this highly practical, full-day workshop, we will explore customization options, execute multi-step changes with Agent Mode, and connect external tools via the Model Context Protocol (MCP).

Participants will learn to delegate issues to the GitHub Copilot coding agent (issue-to-PR), accelerate quality with Copilot code reviews, and work efficiently from the terminal using the Copilot CLI. Through multiple hands-on labs (HOLs), you will implement features, refactor safely, and run review cycles using repeatable templates designed for immediate team adoption.

Workshop Agenda

• Welcome & Environment Setup
• Copilot Fundamentals at Scale: Prompting, verification, and safe workflows.
• Customization Options: Repo instructions, prompt files, and agent-specific guidance.
• Hands-On Labs: Experimenting with the tooling in a live environment.
• Agent Mode: Executing complex, multi-step changes.
• Model Context Protocol (MCP): Connecting Copilot to external tools and context.
• Automation & Review: GitHub Copilot coding agent (issue-to-PR) and automated code reviews.
• GitHub Copilot CLI: Terminal-centric assistance for debugging and execution.
• Bonus: Building custom MCP servers.
• Wrap-up & Next Steps