# ag2ai/ag2

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

Repository: https://github.com/ag2ai/ag2
Canonical: https://ross.abutalabs.com/products/ag2
Homepage: https://ag2.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: a2a, agent-framework, llm, python, agentic, ai, llms, mcp, multi-agent, multi-agent-system, genai, agentic-ai, ai-agents-framework, aiagents, open-source, ag2
Last push: 2026-08-25T19:16:04+00:00

## Health v2 (maintenance only)
Score: 84/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 86, longevity 47
- inputs: {"age_days": 660, "days_push": 8, "days_rel": 18, "gap_med": 7.0, "n_releases_24m": 63}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4890, forks 708 (observed 2026-08-28T04:09:01.990697+00:00)

## What it is
AG2 (formerly AutoGen) is an open-source Python framework for building AI agents and orchestrating multi-agent systems, supporting LLM integration, tool use, human-in-the-loop workflows, and protocols like A2A and MCP. It positions itself as an 'AgentOS' runtime where agents from different frameworks can cooperate.

## Use cases
- build multi-agent ai systems in python
- orchestrate llm agents to solve tasks together
- add tool use and mcp support to ai agents
- create human-in-the-loop agent workflows
- connect agents from different frameworks like langchain and openai
- prototype agentic ai research

## When to choose
- you need multi-agent conversation and coordination patterns
- you want framework interoperability via A2A and MCP protocols
- you want the AutoGen lineage with active maintenance
- you need both autonomous and human-in-the-loop workflows

## When to avoid
- you rely on the classic `import autogen` API, which moved to ag2-classic
- you need a single-agent simple chain rather than agent orchestration
- you want a framework with first-class non-Python support

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, mcp, chatbot, workflow-automation
- domain: large-language-models, artificial-intelligence, developer-tools
- platform: python, cross-platform
- tags: multi-agent, autogen, a2a, agentic-ai, human-in-the-loop, agentos, ai-agents

## Member repositories
- ag2ai/ag2 (main) score 84

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:01.990697+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:18:11.709137+00:00, confidence not recorded.
  - readme: https://github.com/ag2ai/ag2 (fetched 2026-08-28T04:09:01.990697+00:00, sha e687e2a9e9ae)
  - homepage: https://ag2.ai (fetched 2026-08-29T09:00:27.288527+00:00, sha 0afa8ed95a5b)
  - registry_pypi: https://pypi.org/pypi/ag2/json (fetched 2026-08-29T09:00:27.293022+00:00, sha 513b48f49f86)
- Data as of 2026-08-30T08:39:29.467469+00:00.