AI-242: Using GitHub Copilot and spec-kit for Agentic Coding and SDD

Master GitHub Copilot's advanced chat modes, MCP tooling, and the Specification-Driven Development workflow for sustainable software delivery.

Course Length

2 days or 4 half days

Course Overview

AI-powered coding tools such as GitHub Copilot are rapidly transforming how software is designed, implemented, and maintained. This course teaches participants how to move beyond simple prompt-based coding and adopt a systematic, agentic approach that improves productivity without sacrificing quality, governance, or maintainability.

Participants will learn how to use Copilot's advanced chat and agent modes without or with spec-kit to implement “vibe coding” and Specification-Driven Development (SDD) respectively. SDD's structured methodology spans the whole software development life cycle (SDLC) and ensures that software development remains aligned with business intent, technical constraints, and long-term sustainability. The course combines conceptual foundations with extensive hands-on exercises, enabling participants to immediately apply these techniques to real-world projects.

Learning Objectives

By the end of this training, participants will be able to:

  • • Understand the differences between the various modes in Copilot Chat
  • • Understand the concepts of MCP servers and tools, and configure and use them in Copilot
  • • Create new applications and modify existing ones consisting of multiple files
  • • Use Copilot to create and run unit and integration tests
  • • Generate documentation for applications based on source code and user prompts
  • • Use Copilot for source control tasks
  • • Understand the limitations of “vibe-coding” and the need for a more structured methodology which spans the whole software development life cycle (SDLC)
  • • Understand the main principles, advantages and disadvantages of Specification-Driven Development (SDD)
  • • Explain the meaning and importance of SDD steps
  • • Use SDD in greenfield projects
  • • Use SDD in brownfield projects

Main Topics

GitHub Copilot and Agentic Coding

  • Copilot Chat modes: Ask, Edit, Plan, and Agent
  • Configuring and using MCP tools and toolsets
  • Using Copilot to create and modify multi-file applications
  • Using Copilot to create and run unit and integration tests
  • Using Copilot for documentation
  • Using Copilot for source control
  • Configuring and using asynchronous background and cloud-based coding agents

Specification-Driven Development (SDD)

  • Limitations of “vibe coding” and the need for a more structured methodology which spans the whole software development life cycle (SDLC)
  • What is Specification-Driven Development (SDD)?
  • Installation and basic configuration of spec-kit
  • SDD workflow:
    • Constitution
    • Specification
    • Planning
    • Task breakdown
    • Implementation
  • Applying SDD in greenfield projects
  • Applying SDD in brownfield projects

Structure

Roughly 50% lecture, 50% hands on lab exercises. The lab exercises are executables with Python, JavaScript/TypeScript, Java, C# or C/C++.

Target Audience

Software developers and testers as well as their technical managers who want to understand and use GitHub Copilot and spec-kit in agentic coding and specification-driven software development.

Prerequisites

Experience with VSCode or similar IDEs, experience in using ChatGPT or similar chatbots, basic knowledge of the programming language used during lab exercises.

Course Modules

Module: GitHub Copilot and Agentic Coding

  • Copilot Chat modes: Ask, Edit, Plan, and Agent
  • Configuring and using MCP tools and toolsets
  • Using Copilot to create and modify multi-file applications
  • Using Copilot to create and run unit and integration tests
  • Using Copilot for documentation
  • Using Copilot for source control
  • Configuring and using asynchronous background and cloud-based coding agents

Module: Specification-Driven Development (SDD)

  • Limitations of “vibe coding” and the need for a structured methodology spanning the SDLC
  • What is Specification-Driven Development (SDD)?
  • Installation and basic configuration of spec-kit
  • SDD workflow covering Constitution, Specification, Planning, Task breakdown, and Implementation
  • Applying SDD in greenfield projects
  • Applying SDD in brownfield projects

Part of the ComponentSoft AI Portfolio

This training is part of the AI portfolio of Component Soft which explores essential AI topics, such as:

  • • AI-101: Intro to GenAI with Large Language Model (LLMs) and LLM-based apps.
  • • AI-141: Using GitHub Copilot as coding assistant
  • • AI-161: Using Amazon Q as coding assistant
  • • AI-242: Using GitHub Copilot and spec-kit for agentic coding and spec-driven development (SDD)
  • • AI-262: Using Amazon Kiro for agentic coding and spec-driven development (SDD)
  • • AI-434: GenAI Application Development with LLMs
  • • AI-452: Agentic AI Application Development with LLMs