ag2 and AI-Agentic-Design-Patterns-with-AutoGen

AG2 is the core multi-agent orchestration framework, while the AI-Agentic-Design-Patterns repository is an educational course that teaches architectural patterns and best practices for building applications with AG2—making them complements where one provides the foundational technology and the other teaches how to use it effectively.

Maintenance 25/25
Adoption 21/25
Maturity 25/25
Community 22/25
Maintenance 0/25
Adoption 10/25
Maturity 1/25
Community 21/25
Stars: 4,256
Forks: 556
Downloads: 685,501
Commits (30d): 67
Language: Python
License: Apache-2.0
Stars: 135
Forks: 36
Downloads:
Commits (30d): 0
Language: Jupyter Notebook
License:
No risk flags
No License Stale 6m No Package No Dependents

About ag2

ag2ai/ag2

AG2 (formerly AutoGen): The Open-Source AgentOS. Join us at: https://discord.gg/sNGSwQME3x

Enables multi-agent orchestration with built-in conversation patterns (swarms, group chats, nested chats) and supports tool use, code execution, and structured outputs. Built on ConversableAgent primitives that handle message passing between AI entities, with integration across multiple LLM providers beyond OpenAI. Offers both autonomous workflows and human-in-the-loop modes, currently transitioning to a beta framework architecture targeting v1.0 stability.

About AI-Agentic-Design-Patterns-with-AutoGen

ksm26/AI-Agentic-Design-Patterns-with-AutoGen

Learn to build and customize multi-agent systems using the AutoGen. The course teaches you to implement complex AI applications through agent collaboration and advanced design patterns.

Covers practical design patterns including Reflection (nested agent reviews), Tool Use (function calling for legal moves/code execution), and Planning, with hands-on examples from conversational agents and coding tasks to financial analysis. Built on AutoGen's ConversableAgent framework, enabling agents to communicate through nested chats and dynamically call user-defined functions for task execution. The course materials include working implementations across multiple domains—from chess gameplay with tool constraints to multi-agent financial workflows incorporating human-in-the-loop feedback.

Scores updated daily from GitHub, PyPI, and npm data. How scores work