Agent frameworks & infrastructure
Multi-agent orchestration
6,185 agents · most starred first
Multi-agent orchestration, page 6 of 62
Streamlit UI for crewai | crewai ui
Use Claude to orchestrate agents like OpenClaw
AI goes multiplayer. A self-hosted workspace for people and machine people. Create, code, and work together across desktop, mobile, and web. Open source. MIT licensed.
The Multi-Agent Reasoning framework creates an interactive chatbot where AI agents collaborate via structured reasoning and Swarm Integration for optimal answers. Simulating a team that discusses, debates, and refines responses, it enables complex problem-solving and precise results. Now with Prompt Caching to reduce latency and costs.
基于 Hermes Agent 构建的本地多 Agent 团队协作 Web 系统
An infinite canvas product that allows any number of codex, claude, opencode, pi, and other agents to communicate simultaneously on the canvas
Automatically create a crew and tasks for CrewAI
The middle manager for your agents
Self-hosted teams of AI agents that work 24/7, with scheduling, fault recovery, budget limits, and a live dashboard
A smaller, lighter-weight version of OpenClaw, natively multi-agent, compiles to Rust, and built on the Swarms framework and Swarms ecosystem. One API, unified messaging across Telegram, Discord, and WhatsApp with optional Claude-powered reasoning.
A provider-agnostic scaffolding kit for running structured multi-agent workflows in your codebase.
Make your AI agents work as a team. Orchestrate jobs across Claude, Codex, Cursor, Devin, Pi, local models and more, or run a Software Factory: long plans, proven steps, safe merges. Carries your setup, tracks every cost.
Still struggling with the high threshold or looking for the appropriate baseline? Come here and new starters can also play with your own multi-agents!
osBrain - A general-purpose multi-agent system module written in Python
What if AI agents could form teams as fluidly as humans? Spawn specialists in milliseconds, share discoveries in real-time, review each other's work, and remember everything across sessions. Watch it all unfold in a live mission control UI. Open source.
One CLI to orchestrate them all. Manage a team of AI agents executing tasks in parallel from your terminal using a command line interface to manage them all. Manage a team of artificial intelligence agents who perform tasks in parallel from your terminal or offline on a local or server
Clean implementation of Multi-Agent Reinforcement Learning methods (MADDPG, MATD3, MASAC, MAD4PG) in TensorFlow 2.x
Local subagent orchestration for Codex and Claude, with persistent task state, reusable teams, and interruption recovery.
MARLToolkit: The Multi-Agent Rainforcement Learning Toolkit. Include implementation of MAPPO, MADDPG, QMIX, VDN, COMA, IPPO, QTRAN, MAT...
OODA Loop AI Agents - Observe, Orient, Decide, Act framework for systematic problem-solving. Built by AstroMVP to help startups ship AI products fast 🚀
Resources for our paper: "EvoAgent: Towards Automatic Multi-Agent Generation via Evolutionary Algorithms"
Codes accompanying the paper "ROMA: Multi-Agent Reinforcement Learning with Emergent Roles" (ICML 2020 https://arxiv.org/abs/2003.08039)
Private messaging for your AI agents. Free to self-host on Cloudflare. Direct chats, groups and a durable inbox.
31 位思想家组成的多 Agent 审议系统,覆盖工程、商业、人生抉择、关系、心理、创作六大领域。黑格尔正反合结构,一个命令开启深度辩证。
[ICML-25] AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML
Framework and toolkits for building and evaluating collaborative agents that can work together with humans.
An open-source studio for teams of AI agents, from marketing squads to dev teams to research crews. Deploy a team, then watch them delegate and collaborate live. Self-hosted, one command.
A swarm of autonomous AI minds, unified and steered by a single consciousness - the Veil. Runs on any language model.
A cartoon 3D office where a swarm of Claude Code agents (plus QA testers) works through your GitHub issues
Automate complex business workflows with our Multi-AI-Agent Systems using crewAI. This framework leverages autonomous, role-specific AI agents to collaboratively perform multi-step tasks, enhancing efficiency and accuracy across various domains. Ideal for applications in resume tailoring, website design, research, customer support, and more.
Master the art of designing and organizing AI agents. Learn to automate complex, multi-step business processes by creating specialized AI agent teams using the open-source library crewAI.
A cross-platform high-performance native client for Multi Agent Systems
A Peer-to-Peer Autonomous Agent System
Neuro AI System of Agent Networks
Simple, flexible workflow orchestration for multi-agent AI apps, with YAML configuration, runtime safety, observability, and resume support.
Orchestrates multi-step AI agent workflows — defines execution graphs, manages shared state, dispatches work to agents, and tracks progress with checkpoints and streaming
Multi-agent banking assistant with Java and Langchain4j
Self-improving Agent orchestrator for all knowledge work
Mozaik is a TypeScript framework for real-time agent collaboration.
Agent skill that orchestrates Codex, Claude Code, Antigravity, Kimi and custom CLI workers with playbooks that save context and quota, and reports what each run did and cost.
Multi-agent collaboration platform for persistent, proactive AI agents across rooms, workspaces, skills, and external services.
PantheonRL is a package for training and testing multi-agent reinforcement learning environments. PantheonRL supports cross-play, fine-tuning, ad-hoc coordination, and more.
CoordClaw: a multi-agent system that runs AI like a one-person company. Built on management & organizational theory — teams defined in natural language, collaboration driven by a message loop (not hard-coded DAGs). Auditable work logs, human god's-eye oversight.
Cloud-Side Beverage Shop Smart Order Assistant - A distributed multi-agent system built using Spring AI Alibaba Agentic API
MiroFish的免费本地运行版本 | Graphiti+Neo4j替代付费Zep | 简易便捷本地跑通
First-tree routes work to the right agent, gives it the same context your team has, and loops humans in only when the rules say so. Lives in your GitHub. Open source.
Route complex Codex tasks to model-specific background workers with bounded concurrency and lead-agent verification.
Agent-first task orchestration. Persistent memory. Multi-provider coordination.
Claude Multi-Agent Project Manager — multi-channel orchestration, GitHub-first SDK mode, and plugin system for Claude
Graph Engineering, simplified — with GraphCode. #graphcode
把Claude Code的UltraCode模式带给DSH,把 DSH 的一次性多 Agent 调度,升级为可生成、可保存、可治理、可观察、可恢复的 Workflow 层
Multi-model council for DeepSeek Harness: independent answers, anonymous peer reviews, and an inspectable final decision.
Multi-agent orchestration system using Claude Code with Git worktree isolation and specialized AI agents for collaborative development
一个基于多智能体协作的 AI 助手,支持命令行和Web界面,提供团队模式、自定义会议模式和多智能体讨论模式(圆桌会议模式)三种工作方式,通过多个专业智能体协同工作来完成复杂的编程和系统任务。
Open-source multi-agent orchestration harness for OpenCode — specialized agents, durable workflows, research, planning, implementation, review, and validation.
Autonomous AI agent team for one-man companies. Context engineering + harness engineering drive a pipeline that brainstorms, builds, reviews, and ships.
Orchestrate fleets of Claude Code & Claude Computer Use agents across containers, VMs, and physical devices. Live desktop streaming, intelligent task delegation, and multi-channel human-in-the-loop gateway. Schedule tasks with self-healing and self-learning.
A CLI/TUI app that orchestrates multi-agent conversations by enabling different AI CLI tools (Claude Code, Gemini, Qwen, etc.) to communicate in shared rooms.
A dual-agent framework leveraging left-brain logic (structured coding, syntax validation, debugging) and right-brain intuition (macro architecture, creative patterns, UX flows). It splits cognitive loads to boost LLM development efficiency, preventing hallucination while maximizing system scalability.
Elkar - Task management for AI Agents, ready for Agent2Agent (A2A) protocol
Your agent swarm as a live DAG: an orchestration graph with attempts, review gates, and evidence, in a herdr split pane.
Multi-model orchestration subagents for Claude Code — delegates by task scope to Gemini's 1M-token context or GPT's fast iteration, then routes every result back through Claude review.
Multi-Agent Conversation Framework in TypeScript
From isolated agents to shared work
A reference operating model for organizations where any role can be filled by a human or an agent, every process is built for agents first, and every flow can be paused, adjusted by a human, and resumed.
A multi-agent engine for data distillation and long evolutionary collaborative reasoning.
We present the first systematic study on the scaling property of raw agents instantiated by LLMs. We find that performance scales with the increase in the number of agents, using the simple(st) way of sampling and voting. Our method is called Agent Forest, as a tribute to the classic Random Forest.
Agentic Architectural Patterns for Building Multi-Agent Systems, published by Packt
English translation of MiroFish (github.com/666ghj/MiroFish) - A concise and universal swarm intelligence engine for predicting everything.
Turn Fable or Opus into your agent orchestrator: plan coding work, assign capable Claude, Codex, or Grok agents, and personally verify the result.
A persistent, reviewed multi-agent runtime for long-horizon research and engineering.
Mycelium is a shared workspace for AI coding agents and the people who run them.
Swarm.js is a Node.js implementation of OpenAI’s experimental Swarm framework. This SDK allows developers to orchestrate multi-agent systems using OpenAI’s API in a lightweight and ergonomic way, while leveraging Node.js for building scalable, real-world applications.
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.
A multi-agent research laboratory for data distillation and long evolutionary collaborative reasoning.
Pilot Protocol: The Internet of Agents
Gliding Horse is a multi-agent orchestration framework built in Rust that supports PDCA scheduling and knowledge graph-based agents, with comprehensive Chinese documentation, and is suitable for building enterprise-level AI agent systems.
Temporal-native Durable Multi Agent Harness
Instant messaging between coding agent sessions — Claude Code, Codex, opencode, Cursor CLI, MiniMax Code and ZCode — across tools, to other people's agents
Evoplex is a fast, robust and extensible platform for developing agent-based models and multi-agent systems on networks. It's available for Windows, Linux and macOS.
Dynamic decentralized task allocation algorithms for multi-agent systems using auctions
Multi-agent collaboration plugin for Claude Code - orchestrate multiple AI agents (Codex CLI, Gemini CLI, etc.) for diverse perspectives
Messages between AI coding sessions in different vendors' clients, such as Claude Code and Codex. Local, open source, no lead agent. https://getacc.dev
Self-Evolving Multi-Agent Operation System
Production-Grade Multi-Agent Orchestration Framework
**⚔️ Alicization Town** is a decentralized, multi-agent pixel sandbox world powered by the **Model Context Protocol (MCP)**. **⚔️ Alicization Town** 是一个基于 **MCP (Model Context Protocol)** 架构的去中心化多智能体像素沙盒世界。
A coordination protocol for trees of Claude Code agents
Lux is an open-source framework for building multi-agent, swarmed intelligence built by Spectral Labs. With Beams, Prisms, Lenses, and Signals, it orchestrates AI-driven workflows to create truly autonomous organizations.
An open blueprint for intelligent warehouse operations, combining agentic AI, governed warehouse state, decision intelligence, and controlled execution. Agents analyze & propose actions, while deterministic policy, approval, validation, execution layers preserve safety, authority, & auditability.
A multi agents meeting CLI.
Medieval village economy simulation — up to 1000 LLM agents across multiple villages, driven by supply chains, hunger, tool degradation and market pressure.
器灵 Soulmate — 多 AI CLI 编排与社区运营入口
It takes a village to raise a child: Google DeepThink 🧠 but in LangGraph and free - an original algorithm for collaborative agents using qualitative backpropagation. Turn hours of prompting into a single high level prompt and let a QNN figure system prompts on the fly.
Official Implementation of 'UPDeT: Universal Multi-agent Reinforcement Learning via Policy Decoupling with Transformers' ICLR 2021(spotlight)
A visual manager for fleets of Claude Code, Codex and Muse agents — real terminals, agents that message each other, teams that run tickets on worktrees, across your Mac and any Linux box.
[ICML 2026] A neuro-symbolic framework for abductive reasoning that grounds multi-agent collaboration in structured belief states using a causal graph and state machine.