Skip to content

01_Prerequisites.md

Latest commit

 

History

History
284 lines (196 loc) · 8.98 KB

01_Prerequisites.md

File metadata and controls

284 lines (196 loc) · 8.98 KB

Module I — Prerequisites and Environment Setup

Note

Duration: ~10 minutes Ensure all tools and accounts are configured before proceeding to the next module.

Objectives

  • Verify the pre-provisioned Azure resources are accessible
  • Install and verify all required tools
  • Authenticate with Azure using the Azure CLI
  • Confirm GitHub Copilot access in VS Code
  • Validate the Foundry Toolkit for VS Code extension is installed

Step 0: Azure Infrastructure

This workshop requires the following Azure resources. These are provisioned as part of the workshop environment — you do not need to create them from scratch. Module II walks through each resource in detail.

Resource Name Used in Lab Tier
Azure AI Foundry Hub + Project Compliance-Sentinel Standard
GPT-4o model deployment gpt-4o Standard (pay-per-use)
Embedding model deployment compliance-embedding (text-embedding-ada-002) Standard (pay-per-use)
Azure AI Search compliancesearch Free tier
Storage Account + Blob Container compliancekb / kb-documents Standard LRS

Tip

If you are the workshop facilitator, these resources are provisioned by following Module II: Provisioning Azure Resources in advance. Participants can then verify access and proceed directly to the agent design steps.

Verify Azure access:

az login
az account show --query "{Name:name, SubscriptionId:id}" -o table

Step 1: Required Software

Ensure the following software is installed on your machine. If any tool is missing, follow the installation link below.

Tool Version Installation Link
Visual Studio Code Latest Download VS Code
Python 3.10+ Download Python
Azure CLI 2.60+ Install Azure CLI
Git Latest Install Git
Docker Desktop (Optional — Module VII only) Latest Install Docker

Verify Installations

Open a terminal and run the following commands to verify each tool:

# Check Python version
python --version

# Check Azure CLI version
az version

# Check Git version
git --version

# Check Docker version (optional — needed for Module VII)
docker --version

Expected output example:

Python 3.11.7
azure-cli 2.67.0
git version 2.44.0.windows.1
Docker version 27.5.1

Important

Python 3.10 or higher is required. If you have multiple Python versions installed, ensure python --version returns 3.10+. On some systems, you may need to use python3 instead of python.

Step 2: VS Code Extensions

Install the following extensions in Visual Studio Code:

Required Extensions

  1. GitHub Copilot — AI-powered code completions and chat

    • Open VS Code → Extensions panel (Ctrl+Shift+X)
    • Search for GitHub Copilot
    • Click Install

  2. GitHub Copilot Chat — Conversational AI assistant

    • Search for GitHub Copilot Chat
    • Click Install

  3. Foundry Toolkit for VS Code — Agent builder and model playground

  4. Python — Python language support

    • Search for Python
    • Click Install
    • Publisher: Microsoft

Verify Extensions

After installation, you should see:

  • GitHub Copilot icon in the bottom-right status bar
  • Foundry Toolkit for VS Code icon in the left Activity Bar (looks like a chip/brain icon)

Tip

After installation, verify the GitHub Copilot icon appears in the bottom-right status bar and the Foundry Toolkit for VS Code icon appears in the left Activity Bar. If you do not see the Foundry Toolkit for VS Code icon, right-click on the Activity Bar and enable Foundry Toolkit for VS Code.

Step 3: Authenticate with Azure

The workshop requires an active Azure subscription with permissions to create resources.

  1. Open a terminal in VS Code (Ctrl+`` or Terminal → New Terminal).

  2. Sign in to Azure:

    az login

    This will open a browser window. Sign in with your Azure account credentials.

  3. After successful login, verify your subscription:

    az account show --query "{Name:name, SubscriptionId:id, TenantId:tenantId}" -o table

    Expected output:

    Name                                    SubscriptionId                        TenantId
--------------------------------------  ------------------------------------  ------------------------------------
Visual Studio Enterprise Subscription   xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx

  4. If you have multiple subscriptions, set the correct one:

    az account set --subscription "<Your Subscription Name>"

Important

If az login opens a device-code flow instead, follow the on-screen instructions to complete authentication:

az login --use-device-code

Step 4: Sign In to GitHub Copilot

  1. Open GitHub Copilot Chat in VS Code:

    • Press Ctrl+Alt+I to open the Copilot Chat panel
    • Or click the Copilot icon in the Activity Bar

  2. If prompted, sign in with your GitHub account that has an active GitHub Copilot subscription.

  3. Verify Copilot is active:

    • Type a test message in the chat: Hello, are you working?
    • You should receive a response from GitHub Copilot

Note

GitHub Copilot requires one of the following:

  • GitHub Copilot Individual subscription
  • GitHub Copilot Business subscription (managed by your organization)
  • GitHub Copilot Enterprise subscription

If you do not have a subscription, you can start a free trial at github.com/features/copilot.

Step 5: Sign In to Foundry Toolkit for VS Code

  1. Click the Foundry Toolkit for VS Code icon in the VS Code Activity Bar (left sidebar).

  2. You will see the Foundry Toolkit for VS Code panel with sections like:

    • Local Resources
    • Compliance-Sentinel (your Foundry project — will appear after Module II)

  3. Click Sign in to Azure if prompted, and select the same Azure account used in Step 3.

  4. After signing in, you should see your Foundry project and connected resources.

After signing in, expand the Foundry Toolkit for VS Code panel — you should see your Foundry project (e.g., Compliance-Sentinel) with sections for Models, Prompt Agents, Workflows, Tools, and Knowledge.

Step 6: Clone the Workshop Repository

  1. Open a terminal in VS Code.

  2. Navigate to your preferred working directory:

    cd C:\Users\%USERNAME%\Favorites

  3. Clone the repository:

    git clone https://github.com/ADKWDHQWQHDQI/Agent-using-GHCP.git

  4. Open the cloned repository in VS Code:

    cd Agent-using-GHCP
code .

Tip

If you already have the workspace open, you can skip this step.

Step 7: Verify the Knowledge Base Documents

The repository includes 12 compliance documents in the kb_markdown/ folder. Verify they are present:

ls kb_markdown/

Expected output:

DPDP_Act_India_2023.md
EU_China_Data_Flows_2024.md
Export_Control_US_EU_India_2025.md
GDPR_Article_44_Transfers.md
GDPR_Schrems_II_Post_2020.md
Incident_Response_Best_Practices.md
Mitigation_Templates_DPA.md
RBI_Cross_Border_Transactions_Circular_2023.md
RBI_Data_Localization_2018_Guidelines.md
RBI_Vendor_Onboarding_Risk_2024.md
Risk_Scoring_Framework.md
SEBI_Insider_Trading_Policy.md

These documents will be uploaded to Azure Blob Storage in Module II.

Prerequisites Checklist

Before moving to Module II, confirm:

  • Azure resources accessible (AI Foundry, AI Search, Storage Account)
  • Python 3.10+ installed and verified
  • Azure CLI installed and authenticated (az login)
  • VS Code installed with all required extensions
  • GitHub Copilot signed in and responding
  • Foundry Toolkit for VS Code extension installed and signed in with Azure
  • Workshop repository cloned and opened in VS Code
  • 12 KB Markdown documents present in kb_markdown/

Key Takeaways

  • A properly configured environment is essential before building AI agents.
  • Azure CLI authentication enables programmatic access to Azure resources from your local machine.
  • The Foundry Toolkit for VS Code extension bridges VS Code with Azure AI Foundry, allowing you to design and manage agents visually.
  • GitHub Copilot Agent mode will be your primary tool for coding, debugging, and extending the compliance agent.

Click Next to proceed to Module II: Provisioning Azure Resources.