This page provides step-by-step instructions for setting up your development environment to work with the AI Agents for Beginners course. You will learn how to install necessary dependencies, configure environment variables, and prepare your system to run the code examples.
Sources: requirements.txt .env.example .devcontainer/devcontainer.json
Before beginning, ensure you have the following:
AzureCliCredential translations/ja/00-course-setup/README.md125gpt-4o) translations/ja/00-course-setup/README.md147-154Sources: translations/ja/00-course-setup/README.md94-127 02-explore-agentic-frameworks/code_samples/02-semantic-kernel.ipynb57-64
Sources: .devcontainer/devcontainer.json1-17 .devcontainer/Dockerfile1-8 requirements.txt1-10 02-explore-agentic-frameworks/code_samples/02-python-agent-framework.ipynb101-105 02-explore-agentic-frameworks/code_samples/02-semantic-kernel.ipynb118-128
Clone the repository:
Note: A shallow clone is recommended to reduce download size translations/ja/00-course-setup/README.md25-33
Create and activate a virtual environment:
Install dependencies:
Sources: translations/ja/00-course-setup/README.md101-118 translations/ja/00-course-setup/README.md133-135
The repository provides a pre-configured environment using the mcr.microsoft.com/devcontainers/python:3.12 image .devcontainer/Dockerfile1
azure-cli, node lts, and dotnet 10.0 .devcontainer/devcontainer.json4-10ms-python.python, ms-toolsai.jupyter, and ms-dotnettools.dotnet-interactive-vscode .devcontainer/devcontainer.json23-27pip install -r requirements.txt .devcontainer/devcontainer.json16Sources: .devcontainer/devcontainer.json1-30 .devcontainer/Dockerfile1-8
Key libraries used throughout the course:
| Category | Package | Purpose |
|---|---|---|
| Frameworks | agent-framework | Microsoft Agent Framework (MAF) requirements.txt8 |
semantic-kernel | Enterprise AI orchestration 02-explore-agentic-frameworks/code_samples/02-semantic-kernel.ipynb20 | |
| Azure | azure-ai-projects | Azure AI Foundry integration requirements.txt3 |
azure-ai-inference | Unified AI model access requirements.txt2 | |
| Identity | azure-identity | Secure credential management requirements.txt4 |
| Utilities | python-dotenv | Configuration management requirements.txt24 |
| Security | pynacl, jcs | Cryptographic receipts for auditing requirements.txt28-29 |
Sources: requirements.txt1-30
Sources: translations/ja/00-course-setup/README.md166-181 02-explore-agentic-frameworks/code_samples/02-python-agent-framework.ipynb56-59 02-explore-agentic-frameworks/code_samples/02-python-agent-framework.ipynb101-105
Initialize .env:
Copy .env.example to .env .env.example1-26
Azure AI Foundry (Standard Lessons):
AZURE_AI_PROJECT_ENDPOINT: Found in the Project Overview .env.example3AZURE_AI_MODEL_DEPLOYMENT_NAME: The name of your deployed model (e.g., gpt-4o) .env.example6GitHub Models (Lesson 06 & 08):
GITHUB_TOKEN: Your Personal Access Token .env.example13GITHUB_ENDPOINT: https://models.github.ai/inference .env.example14Azure AI Search (Lesson 05 - RAG):
AZURE_SEARCH_SERVICE_ENDPOINT and AZURE_SEARCH_API_KEY .env.example9-10Sources: .env.example1-26 translations/ja/00-course-setup/README.md194-219
The course emphasizes security by using Keyless Connections via AzureCliCredential translations/ja/00-course-setup/README.md192
az login in your terminal translations/ja/00-course-setup/README.md174-17602-explore-agentic-frameworks/code_samples/02-python-agent-framework.ipynb101-105
Sources: 02-explore-agentic-frameworks/code_samples/02-python-agent-framework.ipynb56-105 translations/ja/00-course-setup/README.md166-192
To verify the installation, you can run a simple completion test using the ChatCompletionsClient from the Azure AI Inference SDK:
06-building-trustworthy-agents/code_samples/06-system-message-framework.ipynb29-35
Sources: 06-building-trustworthy-agents/code_samples/06-system-message-framework.ipynb9-35
After completing your environment setup, proceed to Section 2: AI Agent Fundamentals to understand the architectural building blocks of agentic systems.
Refresh this wiki