Dev environment, Copilot access & repo overview

Before you start: use your own account and fork. Sign in with your own GitHub account with active GitHub Copilot access and fork ABBARNABIL/agentic-sdlc-workshop into your account or an organization you control. No accounts or lab credentials are supplied. All later commits, issues, pull requests, Actions runs, and OIDC subjects must target your actual fork. For Challenges 5 and 6, you will use your own Azure subscription and create your own workshop resources. See Start here.

Standalone Codespaces setup. In your fork, use Code → Codespaces → Create codespace on main. This repository has one root dev container, named Agentic SDLC; the upstream monorepo's “Azure / AI / Agentic SDLC” picker is not needed here. The workspace opens at the repository root. Run app commands from src/; the website/ directory builds this guide, not the exercise app.

Goal#

Configure your development environment, confirm GitHub Copilot access, and get familiar with the Octocat Supply application so the later challenges run without setup interruptions.

The Octocat team is behind schedule and adopting Agentic Development practices. Before you can help deliver features, you need a working environment, Copilot enabled, and a clear picture of the codebase.

Actions#

• Get the code — Open your own fork of this repository in GitHub Codespaces (recommended), a local VS Code Dev Container (.devcontainer/), or locally. Confirm you're on the intended working branch.
• Verify runtimes — Confirm the runtimes for your chosen API track are available (TypeScript, Python, .NET, or Java) plus Node.js for the frontend. In Codespaces / the Dev Container these are preinstalled.
• Enable GitHub Copilot — Sign in and confirm your Copilot entitlement is active in your IDE; install the GitHub Copilot / Copilot Chat extensions.
• Explore the repo — Review the structure: src/ (the app, with src/docs/ architecture guidance), infra/ + scripts/ (Azure setup and deployment), assets/ (Challenge 3 backlog & WorkIQ inputs), and .github/ (Copilot custom instructions & agents). Identify where the frontend, APIs, and database live. Read src/README.md for the code structure and build commands.
• Run the app locally — Start the frontend and one API to confirm the application builds and runs before you begin changing it.

Success criteria#

• Your own fork is open in your environment under your GitHub account, and Copilot is active and responding in your IDE.
• You can build and run the frontend and at least one API locally.
• You can describe, at a high level, how the frontend, APIs, and database fit together.

Optional stretch — repo-wide agent context#

Give Copilot persistent, repo-wide context so later challenges start with your conventions loaded.

• Create an AGENTS.md and/or .github/copilot-instructions.md recording coding conventions, naming, and preferred patterns.
• Record the exact build and test commands for your API track (and the frontend) so an agent can verify its own work.
• Add explicit scope rules (e.g. don't touch generated files, infra, unrelated services) to keep agent diffs minimal.
• In a fresh Copilot Chat, ask a question that relies on those conventions without restating them and confirm they're applied.

Learning resources#

• Setting up GitHub Copilot
• GitHub Codespaces
• Development Containers
• Adding repository custom instructions for Copilot