Module 0: Environment & Claude Code Setup

Overview

Before you can build or secure anything, you need a working environment. This module ensures your GCP project, GitHub CLI, AIRS credentials, and Claude Code are all configured and talking to each other. Think of this as the pre-flight checklist before takeoff.

Interactive Lab

The full interactive experience for this module runs in Claude Code. Use /lab:module 0 to begin the guided walkthrough with your AI mentor.

Objectives

• Clone the lab repository and orient yourself to the codebase
• Verify GCP authentication and project configuration
• Verify GitHub CLI authentication for workflow management
• Configure AIRS scanning credentials as GitHub repository secrets
• Understand how CLAUDE.md shapes Claude Code's behavior as your lab mentor

Time Estimate

~30 minutes

Challenges

0.1: Repo & Branch Orientation

You already have a private repo created from the template. Before touching anything, orient yourself to the codebase -- a 3-gate MLOps pipeline that trains, publishes, and deploys a cybersecurity advisor chatbot. Use Claude to build a mental map of where things live and how they connect.

Key areas to explore: workflows (.github/workflows/), model training code (model-tuning/), AIRS scanning (airs/), and the serving application (src/).

Workshop attendees: You'll also set up an upstream remote and sync your git history with the template so you can receive instructor updates and submit PRs.

0.2: Verify GCP Authentication

Confirm that gcloud is authenticated to your lab project, the correct project is set as default, and you can access Cloud Storage.

0.3: Verify GitHub CLI

Verify that the gh CLI is authenticated and can see your repository and its workflows.

0.4: Configure AIRS Secrets

Set three GitHub repository secrets that the scanning pipeline will use: MODEL_SECURITY_CLIENT_ID, MODEL_SECURITY_CLIENT_SECRET, and TSG_ID. These come from your Prisma AIRS tenant.

0.5: Meet Your Assistant

Read CLAUDE.md to understand how Claude has been configured for this lab -- its role as a Socratic mentor, pacing rules, available commands, and hint system.

Key Concepts

• GCP Authentication -- Two layers: local CLI auth (for running commands) and Workload Identity Federation (for GitHub Actions to authenticate to GCP)
• GitHub Secrets -- Encrypted environment variables that workflows access for AIRS SDK authentication
• CLAUDE.md -- Configuration file that shapes Claude Code's behavior for domain-specific workflows

Verification

Run /lab:verify-0 in Claude Code to confirm all setup checks pass and submit your progress to the leaderboard.

What's Next

Your environment is ready. Module 1: ML Fundamentals explores HuggingFace, model formats, datasets, and the decisions enterprises face when adopting open-source AI models.