Skip to content

workshop

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

workshop

README.md

GitHub Copilot CLI Workshop

Welcome to this 90-minute instructor-led workshop on GitHub Copilot CLI! In this hands-on session, you'll learn how to leverage the power of AI-assisted development directly from your terminal.

Workshop Overview

GitHub Copilot CLI is a powerful agentic coding assistant that runs directly in your terminal. Unlike traditional IDE-based Copilot experiences, Copilot CLI enables you to work with AI where many developers spend much of their time - the command line. It can explore codebases, make code changes, run commands, and interact with external tools through Model Context Protocol (MCP) servers.

Duration: 90 minutes

Learning Objectives

By the end of this workshop, you will be able to:

  1. Understand custom instructions - Provide context and guidelines to improve code generation quality
  2. Install and configure GitHub Copilot CLI - Set up Copilot CLI in a codespace and authenticate with GitHub
  3. Work with MCP servers - Extend Copilot CLI's capabilities by connecting to external services
  4. Generate and modify code - Leverage AI to create new features from the terminal
  5. Use agent skills - Enhance Copilot's capabilities with specialized skills
  6. Create custom agents - Build specialized agents for specific tasks

Prerequisites

Before attending this workshop, please ensure you have:

  • A GitHub account with an active Copilot Pro, Pro+, Business, or Enterprise subscription
  • Basic familiarity with terminal/command line operations
  • Git installed and configured

The Application: Tailspin Toys

Throughout this workshop, you'll be working with Tailspin Toys, a crowdfunding platform for games with a developer theme. The application features:

  • Backend: Flask API with SQLAlchemy ORM for database interactions
  • Frontend: Astro/Svelte with Tailwind CSS for styling
  • Database: SQLite for local development

This full-stack application provides an excellent playground for exploring Copilot CLI's capabilities across different file types, languages, and development tasks.

Workshop Exercises

Exercise Topic Description
0. Prerequisites Setup Create your repository and codespace
1. Custom Instructions Context Learn how instruction files guide Copilot
2. Installing Copilot CLI Installation Install and authenticate Copilot CLI
3. MCP Servers External Tools Connect to GitHub and other services via MCP
4. Generating Code Code Generation Use plan mode and generate features
5. Agent Skills Skills Enhance Copilot with specialized skills
6. Custom Agents Agents Create and use custom agents
7. Slash Commands CLI Features Explore context, models, and sharing
8. Review Summary Review key concepts and next steps

Getting Started

Start the workshop with Exercise 0: Prerequisites →

Key Copilot CLI Commands

Command Description
copilot Start an interactive Copilot CLI session
/help Display available commands
/clear Clear the conversation history
/mcp add Add an MCP server
/mcp show Show configured MCP servers
/agent Select a custom agent
/compact Toggle compact output mode

Tips for Success

  1. Be specific - The more context you provide, the better the results
  2. Start with exploration - Ask Copilot to explain the codebase before making changes
  3. Iterate - Refine your prompts based on initial results
  4. Trust but verify - Always review generated code before committing
  5. Use instruction files - Leverage .github/copilot-instructions.md for project-wide guidance

Support

  • During the workshop: Raise your hand or use the chat to ask questions
  • After the workshop: Open an issue in this repository

Happy coding with GitHub Copilot CLI! 🚀