From first autocomplete to production-grade AI-assisted workflows.
An open, self-paced course on GitHub Copilot Pro+ in Visual Studio Code, written for experienced developers who have never used Copilot. Clone this repository, work through the guided prompt examples, and apply the templates directly to your own projects — no registration, no LMS, no cost beyond your Copilot subscription.
Requirements: GitHub account with GitHub Copilot Pro+ active · VS Code 1.126+
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 content works on any paid Copilot plan — since the move to usage-based billing (June 2026), Agent mode, chat, and custom instructions are available on all paid plans. Pro+ is recommended for its larger monthly AI Credit allowance and full model access. 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 AI Credits |
| Cost awareness | Make cost-aware decisions about models and modes to minimize AI Credit 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 |
| 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 |
11 progressive modules across 4 levels. Each module is a single theory page with guided prompt examples (ready prompt → expected output → what to observe) and a companion summary.md with key takeaways and a Self-Check. The hands-on project is the Capstone — 8 deliverables produced after Module 11.
| # | Module | Level | Key skill |
|---|---|---|---|
| 01 | Foundations | Beginner | Install, verify, understand all modes, evaluate AI output |
| 02 | Configuration | Beginner | Optimize VS Code and project structure for AI context |
| 03 | Token Optimization | Beginner | Mode/model decision framework, cost-aware workflows |
| 04 | Prompt Engineering | Intermediate | Structured prompts for every coding scenario |
| 05 | Custom Instructions | Intermediate | Persistent guidance at global, project, and path scope |
| 06 | Agents and Role Specialization | Advanced | 10 role-specialized personas with tool permissions and handoffs |
| 07 | Multi-Agent Workflows | Advanced | Orchestrate agents across complex, multi-step tasks |
| 08 | Advanced Features | Expert | Plan mode, AI review, terminal integration, MCP, CI/CD |
| 09 | AI-Friendly Repository Engineering | Expert | AI-friendly project structure, governance, review protocols |
| 10 | Adoption Roadmap | Expert | 7/30/60/90-day personal and team adoption plan |
| 11 | Platform & GitHub.com Integration | Expert | Coding agent, Copilot in github.com, Copilot CLI, desktop app, surface decisions |
| Folder / File | Purpose |
|---|---|
| agents/ | Agent persona reference library — 10 role definitions + handoff prompts |
| capstone/ | Final project — End-to-End Copilot Workflow Integration |
| checklists/ | AI output review, pre-commit, and completion checklists |
| docs/ | Architecture decisions and design reference |
| instructions/ | Custom instruction examples (global, project, path-scoped) |
| modules/ | Learning modules — theory, guided prompt examples, and summaries |
| prompts/ | Reusable prompt library by category |
| templates/ | Authoring templates for all content types |
| CHANGELOG.md | Release history |
| CONTRIBUTING.md | How to contribute |
| COURSE_OVERVIEW.md | Scope, audience, and key outcomes |
| LEARNING_PATH.md | Guided navigation by level and persona |
| SYLLABUS.md | Full 11-module curriculum detail |
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.