Agentic Coding with GitHub Copilot

The Shifting Role of the Developer

The developer's job is changing: away from writing every line by hand and toward defining requirements, orchestrating agents and reviewing the code they produce. This chapter frames that shift, explains what an agentic coding tool is and provides an overview of the available tools.

• From Code Writer to Analyst, Agent Architect and Reviewer
• What is Agentic Coding?
• The agentic loop: reason, use tools, observe, iterate
• Agentic coding tool landscape: GitHub Copilot, Claude Code, OpenAI Codex and Cursor

Introduction to GitHub Copilot and LLMs

This module explains how Large Language Models work, the difference between static training data and dynamic context, and how GitHub Copilot sits as an orchestration layer on top of these models. Learn how this tool enhances developer workflow through intelligent code suggestions while understanding its capabilities and limitations.

• Understanding LLMs
• LLM Static Knowledge vs. Dynamic Knowledge
• GitHub Copilot Licenses
• GitHub Copilot CLI vs Visual Studio (Code)
• Model Selection: Choosing the Best Model for the Job
• Understanding privacy and security considerations
• LAB: Activating GitHub Copilot Free License

Fundamentals of GitHub Copilot

In this module, we focus on the out-of-the-box experience, exploring how to interact with the Copilot Chat, how to use built-in tools, and how to streamline your daily coding tasks.

• Code Completions and Next Edit Suggestions
• Copilot Chat: Inline vs. Panel
• Agent Modes: Ask, Edit, Agent and Plan
• Copilot Agent Types
• Commit message generation and PR descriptions
• Understanding GitHub Copilot Tools
• LAB: Refactoring and documenting legacy code using standard chat features

AI Across the SDLC: Analysis and Design

Discover how GitHub Copilot transforms every phase of the Software Development Lifecycle (SDLC), from analysis through verification. This module focuses on turning fuzzy requirements into a clear, agent-readable document the agent can build on.

• Extracting and clarifying requirements with AI
• Prompt Engineering Specs
• API and architecture design with Copilot
• Architecture Decision Records (ADRs) as steering context
• Managing issues and work items
• LAB: Designing the E-Commerce Platform - specs, API and architecture

Integrations with MCP and CLI Tools

In this module, you'll learn how to extend GitHub Copilot by connecting it to custom tools and external systems. You'll also be introduced to the Model Context Protocol (MCP), a standard way to connect AI models to tools and data sources. In addition, we'll explore how command-line (CLI) tools can be used to enhance AI agents, and compare this approach with MCP-based integrations.

• Universal AI Integrations with the Model Context Protocol (MCP)
• Useful MCP Servers
• MCP Server Security Considerations
• Context Efficiency: MCP vs CLI Tools
• Providing Copilot with CLI Tools
• LAB: Create a modern UI using Copilot and the Playwright CLI

Skills, Agents and Hooks

This module shows how to tailor AI to your team's way of working. You'll learn how to use Custom Instructions to apply shared standards, Skills to standardize recurring workflows, and Specialized Agents that understand your domain and project context.

• Custom Instructions
• Defining specialized Agents
• Defining reusable capabilities and scripts with Skills
• Automating AI Generated Code Clean-Up with Hooks
• When to use Instructions vs. Skills vs. Agents?
• LAB: Building custom Instructions, Skills and Agents

Expand your Dev Team with GitHub Copilot Cloud Agent

GitHub Copilot Cloud Agent can autonomously implement entire features from high-level requirements, allowing you to guide it through an iterative conversation without using an IDE. Copilot also streamlines code reviews with automated pull request feedback and concise PR summaries.

• Introduction to Cloud Agent
• Defining Feature Requirements with GitHub Issues
• Assigning Copilot as a Pull Request Reviewer
• Steering Copilot
• Reviewing Changes with GitHub Codespaces
• GitHub Agentic Workflows: Defining autonomous AI workflows
• Triggering Agentic Workflows from issues, comments and schedules
• LAB: Add GitHub Copilot to your Project's Development Team

AI Across the SDLC: Implementation and Testing

Once the design is ready, this module shows how to let the agent handle the implementation. You'll learn how to use automated tests, code quality checks, and command-line tools to give the agent clear feedback, helping it produce high-quality code instead of making assumptions or guesses.

• Using test suites as guardrails for the agent
• The Test-Driven Development loop with Copilot
• Static analysis, linters and analyzers as feedback signals
• Hooks for Deterministic Agent Steering
• End-to-end UI verification with the Playwright CLI
• LAB: Building and testing the E-Commerce Platform with Copilot

AI Across the SDLC: Reviewing and Verifying AI-Generated Code

AI can generate a lot of code in a short time, but that doesn't mean the code is always correct. This module teaches you how to review and validate AI-generated code, spot bugs and incorrect suggestions, and use automated tools to verify that the code works as intended before it is deployed.

• Reviewing large AI-produced diffs efficiently
• Static analysis and security scanning as objective review signals
• Verifying behavior with tests, Playwright and benchmarking tools
• Steering the agent to produce reviewable, incremental changes
• LAB: Reviewing, verifying and hardening the E-Commerce Platform

Building Custom Applications with GitHub Copilot SDK

This module shows how to bring the power of GitHub Copilot into your own applications. You'll learn how to use the GitHub Copilot SDK to build custom AI-powered workflows, productivity tools, and enterprise solutions. The module covers how to coordinate tools, work with multiple AI models, and integrate MCP servers, allowing you to create intelligent applications that go beyond the IDE and command line.

• Introduction to the GitHub Copilot SDK
• The Copilot Agentic Core and Execution Loop
• Setting up the SDK and Authenticating
• Integrating Custom Tools and MCP Servers Programmatically
• Building Custom GUIs and Task-Specific Agents
• LAB: Building a Custom Automation Tool with the Copilot SDK