Welcome to a hands-on, multi-section module focused on the GitHub Copilot Coding Agent — an autonomous AI developer that works independently in the background to complete tasks across the full software delivery lifecycle. Unlike IDE-based assistant features, the coding agent operates on GitHub itself, handling everything from branch creation and code changes to pull request authoring and iterative review cycles.
This module walks through every major capability of the coding agent available today, giving you practical experience with each feature in a real workflow.
- Who this is for: Developers, DevOps Engineers, Engineering Managers, Platform Engineers, and anyone who wants to scale their team's output with AI.
- What you'll learn: How to assign tasks, track sessions, steer Copilot's work, customize agent behavior, integrate MCP servers, and manage security — all within a GitHub-native workflow.
- What you'll build: A hands-on understanding of Copilot coding agent across the complete software delivery lifecycle: planning → coding → testing → review → merge.
- Prerequisites: GitHub Copilot Pro, Pro+, Business, or Enterprise plan. Sign up for GitHub Copilot.
- Timing: This module can be completed in approximately 60–90 minutes.
By the end of this module, you'll acquire the skills to:
- Assign GitHub issues and pull requests to Copilot as a coding resource.
- Track, steer, and stop Copilot agent sessions from multiple surfaces.
- Review and iterate on Copilot-generated pull requests.
- Customize agent behavior with instructions, agent profiles, MCP servers, hooks, and skills.
- Use Copilot to address security alerts at scale with security campaigns.
- Integrate the coding agent into IDEs, the GitHub CLI, Raycast, and mobile tools.
- About GitHub Copilot coding agent
- Introduction to prompt engineering with GitHub Copilot
- GitHub Copilot coding agent vs. agent mode
- Enable your GitHub Copilot service (Pro, Pro+, Business, or Enterprise plan required for coding agent)
- Fork and then clone this repository locally for the following exercises. You can also run this lab in a GitHub Codespace from your forked repository.
🚀 Ready to start? Begin with Section 1: Assigning Tasks and Tracking Sessions →
| # | Section | What you'll do |
|---|---|---|
| 1 | Assigning Tasks and Tracking Sessions | Assign issues to Copilot, trigger the agent from PRs and the agents panel, track sessions, steer and stop the agent |
| 2 | Custom Instructions and Code Review | Write custom instructions to guide Copilot's behavior, review Copilot-authored PRs, request Copilot as a code reviewer |
| 3 | Advanced Customization | Create custom agent profiles, configure MCP servers, set up hooks, and enable Copilot Memory |
| 4 | Security, SDLC Integration, and AI Models | Understand security protections, run security campaigns, use the GitHub CLI and Raycast, choose AI models |
- GitHub Copilot coding agent — all how-to articles
- About custom agents
- Extending the coding agent with MCP
- About hooks
- About agent skills
- Responsible use of GitHub Copilot coding agent
- Piloting Copilot coding agent in your organization
- GitHub Copilot Trust Center
- GitHub Changelog — Copilot
- Skills exercise: Expand your team with Copilot coding agent