← back
Piloting agents in GitHub Copilot - Christopher Harrison, Microsoft
Takeaway
Treat GitHub Copilot like a pair programmer — readable code, clear instructions and explicit context matter more than clever prompting.
Summary
- Christopher Harrison (Microsoft) runs a hands-on Copilot lab covering code completion, ask mode, edit mode, agent mode (local) and the new Copilot Coding Agent that can be assigned GitHub issues directly.
- Frames Copilot as 'AI pair programmer' — readable code, comments and good project structure feed context just as they'd help a human collaborator; single-letter variables degrade Copilot performance.
- Key principle: 'don't be passive aggressive with Copilot' — give explicit instructions via prompts, instruction files, and clear intent rather than hoping it infers what you meant.
- Demonstrates instruction files and Copilot Coding Agent assigning issues that get autonomously addressed in a sandbox.
- Repeated emphasis on context: code, comments, structure, prompt clarity and specificity all stack to drive Copilot quality.
copilotcode-generationgithub
Original description
The agent capabilities added to GitHub Copilot have enhanced its ability to act as a peer programmer. Copilot can now discover and generate code based on existing standards, run tests, recover from errors, and call tools using Model Context Protocol (MCP). This workshop will guide you through piloting Copilot's agent capabilities and how to best integrate with the most widely adopted AI coding assistant in the world. Key takeaways include: - Understanding how and when to bring agents into your software development workflow - Providing context through the use of custom instructions and prompt files to ensure consistency across your team - Discovering how MCP provides access to an additional set of external tools and capabilities that the agent can use - Configuring Copilot's agentic capabilities to take advantage of your custom MCP server - Recommended best practices to help your responsibly accelerate your development while maintaining code quality and governance About Christopher Harrison Christopher is a long-time geek who's spent the bulk of his career training, supporting and upskilling developers. He's a web developer at heart with passions which span from Python to DevOps to TypeScript to AI. In his current role as an Enterprise Advocate for GitHub he seeks to help organizations improve their DevOps process and culture. When not found writing code he can be found running, playing Civilization, or spending time with his partner and their four-legged child (a rescue mutt). Recorded at the AI Engineer World's Fair in San Francisco. Stay up to date on our upcoming events and content by joining our newsletter here: https://www.ai.engineer/newsletter