Open-source agent runtime and terminal TUI for long-horizon research and file-based work, with ReAct and parallel Agent Team modes.
microsoft/ai-agents-for-beginners
A Microsoft-led course of 18 lessons teaching beginners how to build AI agents, with Python code samples using Microsoft Agent Framework and Foundry.
- Stars
- 77,507
- Forks
- 25,421
- Open issues
- 6
- Last commit
- 3 days ago
What is it?
AI Agents for Beginners is a structured course that walks developers through the core concepts of building AI agents, from the basics of what an agent is to patterns for tool use, agentic RAG, multi-agent collaboration, memory, and security. Each lesson is a self-contained written chapter with code samples, a short video, and links to further reading, so learners can start from any topic.
The code examples are written in Python and use Microsoft Agent Framework together with Microsoft Foundry Agent Service V2, which requires an Azure account. Some samples also work with OpenAI-compatible providers such as MiniMax. Later lessons cover protocols like MCP and A2A, context engineering, computer use agents built on browser automation, deploying scalable agents, and running local agents with Foundry Local.
The course is translated into more than 50 languages, including Simplified Chinese, and is part of a wider set of beginner courses maintained by the same team.
Key features
- 01Covers 18 lessons from agent fundamentals to production deployment and security.
- 02Each lesson includes a written chapter, Python code samples, and a short video.
- 03Code uses Microsoft Agent Framework with Microsoft Foundry Agent Service V2.
- 04Some samples can run with OpenAI-compatible providers.
- 05Available in more than 50 translated languages.
Use cases
- Learning AI agent design patterns before choosing a framework.
- Building a first tool-using or multi-agent prototype in Python.
- Studying how agents are deployed, secured, and run locally.
Quick start
- 1
Clone the repository (the README's sparse checkout skips the large translation folders)
git clone --filter=blob:none --sparse https://github.com/microsoft/ai-agents-for-beginners.git - 2
Enter the project folder
cd ai-agents-for-beginners - 3
Read the Course Setup guide for configuring and running the code samples
Things to know
- The code samples require a Microsoft Foundry account, and the README says an Azure account is required.
Topics
FAQ
- What is ai-agents-for-beginners?
- A Microsoft-led course of 18 lessons teaching beginners how to build AI agents, with Python code samples using Microsoft Agent Framework and Foundry. AI Agents for Beginners is a structured course that walks developers through the core concepts of building AI agents, from the basics of what an agent is to patterns for tool use, agentic RAG, multi-agent collaboration, memory, and security. Each lesson is a self-contained written chapter with code samples, a short video, and links to further reading, so learners can start from any topic.
- Is ai-agents-for-beginners free to use? What license does it have?
- ai-agents-for-beginners is open source under the MIT license. Check the license terms before using it in commercial products.
- Is ai-agents-for-beginners still maintained?
- Actively maintained. The last commit to ai-agents-for-beginners was on Oct 18, 2026.
- What can I use ai-agents-for-beginners for?
- Learning AI agent design patterns before choosing a framework. Building a first tool-using or multi-agent prototype in Python. Studying how agents are deployed, secured, and run locally.
- How do I install ai-agents-for-beginners?
- Clone the repository (the README's sparse checkout skips the large translation folders): git clone --filter=blob:none --sparse https://github.com/microsoft/ai-agents-for-beginners.git Enter the project folder: cd ai-agents-for-beginners Read the Course Setup guide for configuring and running the code samples
- Where can I get ai-agents-for-beginners?
- The source code and installation instructions are on GitHub: https://github.com/microsoft/ai-agents-for-beginners
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A Python library and runnable Jupyter textbook implementing 35 agentic AI architectures, from Reflexion and ReAct to MemGPT and BrowserAgent, on LangGraph.
A collection of 50+ runnable Jupyter notebook tutorials covering GenAI agent techniques, from simple chatbots to multi-agent systems.
A free, open-source 10-week online course on building LLM applications with RAG, vector search, agents, evaluation and monitoring.
An agent-native, self-hostable AI learning workspace for personalized tutoring, Q&A, quizzes, research and visualization, built on Python and Next.js.
Open-source Python framework for building production-ready LLM applications with modular pipelines and agent workflows for RAG, search and agents.