WeAreDevelopers Full-Day Hands-On Lab — Spec-Driven Development: Building & Modernizing with Agentic AI.
This is a MoaW workshop. The best way to read it is rendered:
Or read the raw markdown: docs/workshop.md.
The lab is deliberately open at the agentic-AI-tool level. Hands-on exercises are covered for the three currently dominant surfaces:
- GitHub Copilot in VS Code
- GitHub Copilot CLI
- Claude Code (Anthropic)
- Getting started with agentic coding: completions, chat, agent/plan modes, terminal agents, project-level instructions (
copilot-instructions.md,AGENTS.md,CLAUDE.md), reusable prompts / slash commands, custom agents / subagents, skills and MCP — on all three surfaces. - Spec-Driven Development (SDD): Make specifications, not vibes, drive what the agent builds. End-to-end TypeScript/Node example, portable across Copilot and Claude Code.
- spec-kit: an open-source toolkit by the GitHub team that formalizes the SDD loop across many agentic coding tools.
- (coming) SDD for app modernization: a dedicated hands-on chapter on using SDD to modernize legacy apps.
- A GitHub account with Copilot access (Free, Pro, Business or Enterprise)
- VS Code with the GitHub Copilot + Copilot Chat extensions
- Node.js 20+
- GitHub CLI (
gh) and the GitHub Copilot CLI (copilot) - Optional but recommended: Claude Code (
@anthropic-ai/claude-code) with an Anthropic account that has Claude Code access
See the Pre-requisites section of the workshop for full details.
Based on the jkordick/ghcp-advanced workshop (Chapters 1–3 and the modernization placeholder), extended to be tool-agnostic across GitHub Copilot, the Copilot CLI and Claude Code for the WeAreDevelopers full-day lab.
Originally inspired by the excellent GitHub Copilot HoL by @Philess.
MIT — see LICENSE.