From first autocomplete to production-grade AI-assisted workflows.
An open, self-paced course on GitHub Copilot Pro+ in Visual Studio Code. Clone this repository, follow the labs, and apply the templates directly to your own projects — no registration, no LMS, no cost beyond your Copilot subscription.
- Getting Started
- Who This Course Is For
- What You Will Learn
- Quick Navigation
- Course Structure
- Repository Contents
- Contributing
- Feature Verification
- License
Requirements: GitHub account with GitHub Copilot Pro+ active · VS Code 1.90+
git clone https://github.com/Clipperone/copilot-learning-labs.git
cd copilot-learning-labsOpen the folder in VS Code, install the recommended extensions when prompted, and sign in to GitHub.
Then start here → LEARNING_PATH.md
Note
Most beginner and intermediate content works on any paid Copilot plan. Agent mode and premium model access require Pro+. Plan-specific restrictions are noted at the module level.
| Learner | Starting point | Primary goal |
|---|---|---|
| Developer new to Copilot | No Copilot experience | Get productive fast with a structured foundation |
| Developer using Copilot informally | Uses completions and basic chat | Move from ad hoc usage to deliberate, repeatable workflows |
| Engineering manager | Team has Copilot, no standards | Define team conventions, instructions, and adoption milestones |
| Developer productivity coach | Copilot experience, no training framework | Build and deliver a structured, practical training program |
Prerequisite knowledge: Basic familiarity with VS Code and at least one programming language. No prior Copilot experience required.
| Skill area | What you will be able to do |
|---|---|
| Setup and modes | Configure Copilot Pro+ and VS Code for maximum productivity; choose the right mode — inline completion, inline chat, Ask, Plan, Agent — for any task |
| Prompt engineering | Write effective, repeatable prompts for code generation, refactoring, debugging, testing, documentation, and security review |
| Custom instructions | Design persistent instructions that guide Copilot consistently across a project at global, project, and path scope |
| Agent workflows | Define role-specialized agents with clear responsibilities, tool permissions, and handoff protocols |
| Multi-agent orchestration | Orchestrate agents across complex, multi-step tasks without wasting context or premium requests |
| Cost awareness | Make cost-aware decisions about models and modes to minimize premium request consumption |
| Adoption planning | Apply a structured 7/30/60/90-day personal and team adoption roadmap |
| I want to… | Go to |
|---|---|
| Follow the course from the beginning | LEARNING_PATH.md |
| See all modules and topics at a glance | SYLLABUS.md |
| Start the first lab immediately | labs/lab-01-getting-started/ |
| Find a reusable prompt | prompts/ |
| Read the full course overview | COURSE_OVERVIEW.md |
| Review AI-generated code safely | checklists/ai-output-review.md |
| Understand what was recently added | CHANGELOG.md |
10 progressive modules across 4 levels, each paired with a hands-on lab and a self-assessment checklist.
| # | Module | Level | Lab | Key skill |
|---|---|---|---|---|
| 01 | Foundations | Beginner | Lab 01 — Getting Started | Install, verify, understand all modes, evaluate AI output |
| 02 | Configuration | Beginner | Lab 02 — Project Configuration Baseline | Optimize VS Code and project structure for AI context |
| 03 | Token Optimization | Beginner | Lab 03 — Token Audit | Mode/model decision framework, cost-aware workflows |
| 04 | Prompt Engineering | Intermediate | Lab 04 — Prompt Engineering Workshop | Structured prompts for every coding scenario |
| 05 | Custom Instructions | Intermediate | Lab 05 — Write Your Project's Custom Instructions | Persistent guidance at global, project, and path scope |
| 06 | Agents and Role Specialization | Advanced | Lab 06 — Agents and Personas | 10 role-specialized personas with tool permissions and handoffs |
| 07 | Multi-Agent Workflows | Advanced | Lab 07 — Run a Complete Multi-Agent Workflow | Orchestrate agents across complex, multi-step tasks |
| 08 | Advanced Features | Expert | Lab 08 — Advanced Feature Tour | Plan mode, AI review, terminal integration, CI/CD |
| 09 | AI-Friendly Repository Engineering | Expert | Lab 09 — Repository Health Audit | AI-friendly project structure, governance, review protocols |
| 10 | Adoption Roadmap | Expert | Capstone | 7/30/60/90-day personal and team adoption plan |
| 11 | Platform & GitHub.com Integration | Expert | Capstone (Deliverable 8) | Coding agent, Copilot in github.com, gh copilot CLI, surface decisions |
| Folder / File | Purpose |
|---|---|
| modules/ | Learning modules — theory, exercises, and checklists |
| labs/ | Hands-on labs — starter files and reference solutions |
| prompts/ | Reusable prompt library by category |
| instructions/ | Custom instruction examples (global, project, path-scoped) |
| agents/ | Agent persona definitions — populated during Lab 06 |
| templates/ | Authoring templates for all content types |
| checklists/ | AI output review, pre-commit, and completion checklists |
| docs/ | Architecture decisions and design reference |
| capstone/ | Final project — End-to-End Copilot Workflow Integration |
| COURSE_OVERVIEW.md | Scope, audience, and key outcomes |
| SYLLABUS.md | Full 10-module curriculum detail |
| LEARNING_PATH.md | Guided navigation by level and persona |
| CONTRIBUTING.md | How to contribute |
| CHANGELOG.md | Release history |
Contributions of all kinds are welcome — content fixes, new prompts, improved labs, and translation notes.
Before opening a PR:
- Read CONTRIBUTING.md for conventions and template requirements.
- For significant changes, open an issue first.
- Follow the CODE_OF_CONDUCT.md.
Use GitHub Discussions for questions, learning support, and ideas that are not yet ready for an issue.
This course documents GitHub Copilot features as they exist at publication time. Each module includes a Verified: YYYY-MM date. Copilot evolves quickly — if you find outdated content, open a bug report.
Official references:
MIT — free to use, adapt, and share with attribution.