Hands-on with GitHub Copilot's agents

The recent additions to the capabilities of GitHub Copilot provide powerful tools to the developer across the entire software development lifecycle (SDLC). This includes working with issues and pull requests on GitHub, interacting with external services, and of course code creation. This lab explores the functionality, providing real-world use cases and tips on how to get the most out of the tools.

Because GitHub Copilot is probabilistic rather than deterministic, the exact code, files changed, etc., may vary. As a result, you may notice slight differences between screenshots and code snippets in the lab and your experience. This is to be expected, and is just the nature of working with this class of tools.

If something appears broken or isn’t running correctly, please ask a mentor!

Choose your harness

GitHub Copilot meets you wherever you work. Pick the harness that matches how you want to build, and work through its exercises against a shared Tailspin Toys backlog. Each harness starts with its own setup, so you can dive straight into the one you choose.

🖥️ VS Code

GitHub Copilot inside Visual Studio Code and GitHub Codespaces. Work with Copilot Chat agent mode, MCP servers, and custom agents without leaving the editor you already use — ideal when you want AI assistance woven directly into your IDE.

💻 Copilot CLI

GitHub Copilot CLI — an agentic assistant that runs in your terminal. Install it, connect MCP servers, generate code with plan mode, and build your own skills, custom agents, and slash commands, all from the command line.

🤖 Copilot App

The GitHub Copilot app — a desktop application built on Copilot CLI. Run parallel agent sessions, switch session modes, collaborate on canvases, and manage GitHub issues and pull requests natively — including Agent Merge, which shepherds a pull request through rebases, review feedback, CI fixes, and merge.

☁️ Copilot Cloud Agent

Copilot cloud agent — an asynchronous peer programmer that works on GitHub issues in the background. Assign work, guide it with custom agents, monitor progress from the agents dashboard, and review the pull requests it opens.

Scenario

You are a new developer for Tailspin Toys, a fictional company who provides crowdfunding for board games with a developer theme - a huge market! Your team’s backlog is already filed as GitHub issues, ready for you to pick up — feature work (like filtering and pagination) alongside quality improvements (like accessibility and coding standards). You’ll work iteratively, exploring both the site and Copilot’s capabilities, to complete the tasks.

Get started

Choose your harness above to begin — each one opens with the setup it needs to get you building.