GitHub Copilot: Agentic Engineering Fundamentals (GH-300)

Boost your productivity with GitHub Copilot.

Description

This course introduces GitHub Copilot and its ecosystem through practical labs and demos. Participants will learn how to integrate Copilot into their daily workflows efficiently.

The course starts with the foundations of GitHub Copilot, including installation, access management, security, and compliance.

You'll explore the context mechanisms that drive code completions.

You will work with all major Copilot features: inline suggestions, next edit suggestions, and Copilot Chat (Ask, Agent, and Plan modes).

Practical labs on your own projects reinforce each topic, enabling you to apply the features immediately to realistic use cases.

Customization is a core component. You'll learn how to tailor Copilot behavior using instruction files, prompt files, etc. — the practice of context engineering. Diving into Model Context Protocol (MCP) for integrating custom tools and context-aware extensions into Copilot, building and running a small custom made MCP server of your own.

The course includes best practices for prompt writing, context management, quality assurance, and responsible usage. Security risks, potential model biases, and limitations of generative AI are discussed explicitly.

You'll also deploy a local LLM using Open WebUI to explore privacy-conscious alternatives.

By the end of this course, you will confidently apply GitHub Copilot and its ecosystem to improve code quality, spend less time writing boilerplate, and integrate AI workflows into real-world development tasks.

Note: The Microsoft course GH-300 is a certification-focused course. This Info Support course goes deeper and emphasizes applied usage in realistic projects.

Learning Goals

• Apply code completions using inline suggestions and next edit proposals to accelerate development [Apply]
• Use Copilot Chat in Ask, Agent, and Plan modes to interact with and modify codebases [Apply]
• Describe the strengths, limitations, and ethical considerations of using GitHub Copilot [Understand]
• Customize Copilot using instruction files, prompt files and the Model Context Protocol (MCP) [Apply]
• Explore smart actions like CLI tools and semantic search to improve productivity [Apply]
• Interact with a local large language model (LLM) for improved privacy and control [Understand]

Subjects

• Installing and configuring GitHub Copilot
• Using inline and next edit suggestions
• Copilot Chat: Ask, Agent, and Plan modes
• Copilot in the CLI: explain, suggest, and inline chat
• Customization via instruction files, prompt files, etc.
• Copilot extensions and advanced tools with Model Context Protocol (MCP), including using your own
• Smart actions: semantic search, generate commit messages and Pull Request summaries
• Writing effective prompts and using AI responsibly
• Running and integrating local LLMs using Open WebUI