# inclusionAI/AWorld

Search, understand, reproduce, and improve an idea with ease

Repository: https://github.com/inclusionAI/AWorld
Canonical: https://ross.abutalabs.com/products/aworld
Homepage: https://www.aworldagents.com
Language: Python
License: MIT
License Family: permissive
Topics: world-model, mcp, agent-learning, agent-runtime, environment, agent-framework, browsecomp, gaia, rl-training, xbench
Last push: 2026-09-02T13:34:13+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 73, longevity 38
- inputs: {"age_days": 537, "days_push": 0, "days_rel": 100, "gap_med": 51.5, "n_releases_24m": 7}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1227, forks 126 (observed 2026-09-03T02:15:12.908085+00:00)

## What it is
AWorld is an open-source agent framework and runtime that orchestrates tools, memory, context, and execution for building autonomous AI agents. It lets domain experts encode their knowledge into reusable Skills and deploy fleets of agents, with a CLI for tasks like deep search and app creation.

## Use cases
- build autonomous ai agents with tools and memory
- encode domain expertise into reusable agent skills
- run deep search workflows with a browsing agent
- orchestrate multi-agent systems with mcp tools
- train and evaluate agents on benchmarks like gaia and browsecomp
- create small apps from a single prompt via an agent cli

## When to choose
- you want a full agent harness with runtime, skills, and CLI out of the box
- you need MCP support and multi-agent orchestration in Python
- you want to benchmark agents on GAIA, BrowseComp, or xbench
- you want experts to contribute domain knowledge as reusable skills

## When to avoid
- you only need a simple single-prompt LLM wrapper without tool orchestration
- you need a production web framework rather than an agent runtime
- your stack is not Python-based

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, mcp, llm-inference, rag, cli, workflow-automation
- domain: large-language-models, artificial-intelligence, developer-tools
- platform: python, cli, cross-platform
- tags: agent-runtime, world-model, agent-skills, deep-search, rl-training, multi-agent, agent-harness, ai-agents, automation

## Member repositories
- inclusionAI/AWorld (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:12.908085+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 2026-08-30T06:15:19.723167+00:00, confidence not recorded.
  - readme: https://github.com/inclusionAI/AWorld (fetched 2026-09-03T02:15:12.908085+00:00, sha 18d960b0a7f4)
  - homepage: https://www.aworldagents.com (fetched 2026-08-29T12:23:28.188559+00:00, sha 86aec57d2698)
  - registry_pypi: https://pypi.org/pypi/aworld/json (fetched 2026-08-29T12:23:28.197952+00:00, sha a9c065da68a0)
- Data as of 2026-08-30T08:39:29.467469+00:00.