AI agent framework / autonomous computer-use agent
Agent Zero is an open AI agent framework that gives an LLM a Dockerized Linux desktop with GUI/browser/document tools, projects, memory, plugins, and multi-agent cooperation.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
Choose it for a lightweight terminal-native agent toolkit with a unified LLM API instead of a Dockerized desktop environment.
AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI
Choose it for agent engineering on a TypeScript platform instead of a prebuilt Linux-desktop agent.
AI Agent Engineering Platform built on an Open Source TypeScript AI Agent Framework
Choose it for an extensible agent that operates directly on your host (install/edit/test) rather than inside a Docker desktop.
an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Choose it to build Python agent platforms when you don't need a ready-to-run desktop environment.
Build, run, and manage agent platforms.
Choose it for TypeScript-native agent/LLM application development without an included desktop environment.
Mastra is the modern TypeScript framework for AI-powered applications and agents.
Choose it for stateful agents with persistent memory when long-term agent state matters more than OS control.
Platform for stateful agents: AI with advanced memory that can learn and self-improve over time.
Choose it for Python agent development and evaluation with strong model/skill/MCP support.
Code, Build and Evaluate agents - excellent Model and Skills/MCP/ACP/A2A Support
Choose it when you want a standalone Python AI-assistant framework with skills under AGPL.
An open-source AI assistant framework with skills and agent architecture
Choose it to build Python multi-agent applications when you need framework-level control, not a prebuilt desktop agent.
agentUniverse is a LLM multi-agent framework that allows developers to easily build multi-agent applications.
Choose it for multiplayer/harness-based human-agent work collaboration rather than a single Dockerized desktop agent.
Multiplayer agent harness for work.
Choose it for a lightweight, customizable Rust agent infrastructure when you don't need a full desktop environment.
Lightweight, clear, and fully extensible AI agent infrastructure — learn easily, customize anything 🐉
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
Choose it for structured role-based multi-agent software generation, not for general computer/desktop control.
🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming
Choose it for conversational multi-agent software development, not for full desktop automation.
ChatDev 2.0: Dev All through LLM-powered Multi-Agent Collaboration
Choose it to autonomously plan/code/test production PRs on AgentField, narrower than whole-desktop automation.
Autonomous software engineering fleet of AI agents for production-grade PRs on AgentField: plan, code, test, and ship.
Choose it for a DeepSeek-optimized terminal coding agent, not a general computer-use framework.
DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability — leave it running.
Choose it when you want a Qwen-backed terminal coding agent instead of controlling a full Linux desktop.
An open-source AI coding agent that lives in your terminal.
Choose it for autonomous software development (code/test/PR), a narrower task than whole-desktop operation.
Develop software autonomously.
Choose it for coding-agent help inside an IDE, a coding-only narrower scope than full desktop automation.
open-source coding agent
A previous-generation hierarchical agent-swarm framework, archived and long-unmaintained; use only for legacy designs.
HAAS = Hierarchical Autonomous Agent Swarm - "Resistance is futile!"
Choose it for autonomous red-team security testing, a specialized agent narrower than general desktop automation.
Autonomous Hacking Agent for Red Team
An earlier terminal/browser/editor autonomous agent; unmaintained for over two years, so choose only for legacy use.
✨ Fully autonomous AI Agent that can perform complicated tasks and projects using terminal, browser, and editor.
These projects were analysed and named agent-zero among their alternatives. The relationship is not symmetric — how agent-zero rates them is a separate judgement, made when agent-zero is analysed in its own right.
calls agent-zero “choose it for a Python agent framework with a different extensibility model and built-in Docker execution.”
🧠 Leon is your open-source personal assistant.
calls agent-zero “choose for a lightweight, customisable autonomous agent framework with tool use, if you prefer low-level control over a pre-built platform.”
🙌 OpenHands: AI-Driven Development
calls agent-zero “Choose when you prefer a Python autonomous agent framework with local/human-in-the-loop operation, but verify its license before adopting.”
an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
calls agent-zero “Choose Agent Zero if you want a flexible Python autonomous-agent framework without Letta's managed memory/runtime.”
Platform for stateful agents: AI with advanced memory that can learn and self-improve over time.
calls agent-zero “Choose for a general-purpose personal agent framework that can spawn subordinate agents, with a different runtime and web UI focus.”
agentUniverse is a LLM multi-agent framework that allows developers to easily build multi-agent applications.
calls agent-zero “Python dockerized personal agent app; choose it if you want a ready-to-run autonomous assistant, not an embeddable library.”
An open-source, code-first Typescript toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.