GitHub Copilot in 2026: Chat, Edits, Agents, Spaces, and CLI

Class Duration

14 hours of live training delivered over 2-3 days to accommodate your scheduling needs.

Student Prerequisites

Professional software development experience in at least one language
Active GitHub Copilot subscription (Individual, Business, or Enterprise)
One supported host editor installed: VS Code, Visual Studio, a JetBrains IDE, or another Copilot-supported editor

Target Audience

Software engineers, tech leads, platform engineers, and DevEx teams who want a comprehensive, up-to-date picture of GitHub Copilot's current capabilities—moving well beyond basic code completion into chat, multi-file edits, agentic workflows, Spaces, and MCP server integration. Supersedes earlier single-topic Copilot courses.

Description

GitHub Copilot has evolved into a comprehensive AI development platform, and this course covers the full current feature set. We start with the foundations (setup, model selection, inline suggestions, Next Edit Suggestions with inline previews, context) and move through the complete chat surface (Ask, Edit, Agent modes), multi-file edits, coding agents for longer-horizon tasks, the Copilot CLI for terminal-based agentic workflows, and the Spaces feature for collaborative AI workflows. We also cover deep integration via custom instructions, prompt files, chat modes, MCP servers, and Copilot Extensions—giving teams the tools to standardize how Copilot works across the organization.

Learning Outcomes

Configure GitHub Copilot end-to-end: plan, model selection, IDE setup, extensions.
Use inline suggestions effectively with context awareness and ghost text navigation.
Use Next Edit Suggestions, including inline edit previews and far-away edits.
Drive all three chat modes—Ask, Edit, and Agent—for progressively autonomous tasks.
Manage multi-file edits using Copilot Edits and agent-driven workspace modifications.
Use the Copilot CLI for terminal-based agentic tasks and shell-driven workflows.
Set up and use GitHub Copilot Spaces for collaborative, context-rich AI sessions.
Customize Copilot behavior with instructions files, prompt files, and custom chat modes.
Integrate MCP servers to ground Copilot in internal tools, APIs, and data sources.
Apply code review, commit message, and PR automation features to the full development cycle.
Mitigate privacy, security, and public code match risks.

Training Materials

Comprehensive courseware is distributed online at the start of class. All students receive a downloadable MP4 recording of the training.

Software Requirements

Active GitHub Copilot subscription, a supported host editor with the GitHub Copilot extension installed (VS Code, Visual Studio, or a JetBrains IDE such as IntelliJ IDEA, PyCharm, WebStorm, or Rider), and Git.

Training Topics

Setup and Configuration

Subscription plans and enterprise policy controls
Model selection and switching
Supported host editors: VS Code, Visual Studio, JetBrains IDEs (IntelliJ, PyCharm, WebStorm, Rider, GoLand), Eclipse, Xcode, Vim/Neovim, and Azure Data Studio
IDE installation and feature flags
Privacy settings and content exclusions

Inline Suggestions and Context

Ghost text, accept/reject, and partial acceptance
Next Edit Suggestions: inline edit previews and far-away edits
Context sources: open files, workspace, and imports
Improving suggestion quality with context
Language-specific behaviors

Chat Modes: Ask, Edit, Agent

Ask mode: conversational exploration and Q&A
Edit mode: targeted inline code changes
Agent mode: multi-step autonomous task execution
Choosing the right mode for the task

Multi-File Edits and Copilot Edits

Initiating and reviewing multi-file changes
Edit rollback and selective acceptance
Large refactors with agent assistance

GitHub Copilot CLI

Installation and authentication; April 2026 GA release
Terminal-based agent sessions with named, resumable sessions
Specialized custom agents: Explore (codebase analysis), Task (commands/tests/builds)
Autopilot mode for autonomous, lower-supervision workflows
/fleet: split a task across multiple parallel subagents (CLI-exclusive)
Copilot Spaces in the CLI via the GitHub MCP server
Shell-driven workflows and scripting Copilot from the command line
When to reach for the CLI vs. the IDE agent

GitHub Copilot Spaces

Creating and configuring a Space
Sharing context and prompt packs across a team
Using Spaces for onboarding and knowledge transfer
Space-level instruction and tool configuration

Custom Instructions, Prompt Files, and Chat Modes

Writing .github/copilot-instructions.md
Prompt files for repeatable workflows
Custom chat modes for specialized personas
Team-wide instruction management
Pairing Copilot with GitHub Spec Kit for spec-driven workflows (covered in depth in the dedicated Spec Kit course)

MCP Servers and Extensions

What MCP servers do and how Copilot uses them
Connecting local and remote MCP servers
GitHub Copilot Extensions ecosystem
Building a simple internal MCP tool
Coexisting with the OpenAI Codex IDE extension in the same VS Code workspace

Code Review, Commits, and PRs

Copilot code review in VS Code and on GitHub
Generating commit messages from diffs
PR description and summary automation
Security scanning integration

Workshop

Agent task lab: implement and test a feature end-to-end
MCP integration exercise
Q&A session