Agent frameworks & infrastructure
Multi-agent orchestration
6,185 agents · most starred first
Multi-agent orchestration, page 7 of 62
Orchestrator for AI coding (claude code, cursor, codex, gemini)
An async-first framework for building multi-agent AI systems with an innovative approach to parallelism, so you can focus on creating intelligent agents, not on managing the concurrency of your flows.
Run agents like a company. AgentOS is the native control plane for OpenClaw — manage agents, tasks, models, context, approvals, and runtime visibility from one place.
Open-source implementation of SeaSeed.ai v1.0 (codename: Clawerse). AI ocean world platform for multi-agent social, tasks and compute collaboration.
Agent orchestration & security template featuring MCP tool building, agent2agent workflows, mechanistic interpretability on sleeper agents, and agent integration via CLI wrappers
gradio region:us
Pi multi-agent framework for power users: background/foreground agents, sync/async coordination, fresh/forked context, child-to-parent messaging, orchestrator mode, supports tmux | cmux | ZelliJ | WezTerm | Herdr
Lean implementation of various multi-agent LLM methods, including Iteration of Thought (IoT)
Build & Share AI agents with your team. Full AgentCore, Full Serverless, Full TypeScript Sample
A lightweight, event-driven multi-agent framework for embodied AI systems.
AI-powered cascading development framework. Decompose complex projects into parallel executable tasks with auto-generated PRDs, design docs, and multi-agent collaboration (Claude Code, Codex, Aider).
Trajectory generation and simulation for multi-agent swarm
Multi-agent orchestration CLI — your agents, all together
MAGNet: Multi-agents control using Graph Neural Networks
Self-hosted AI Agents for your whole team — on your infrastructure, your models, around the clock.
AI agent 通用任务治理框架:对齐目标与事实,规划和调度能力,守住授权与风险边界,治理任务执行到真实验收与交付。Governance framework for evidence-driven planning, orchestration, and verified delivery.
A code implementation for our arXiv paper "Multi-agent Adhoc Team Play using Decompositional Q function"
some Multiagent enviroment in 《Multi-agent Reinforcement Learning in Sequential Social Dilemmas》 and 《Value-Decomposition Networks For Cooperative Multi-Agent Learning》
Block-style programming environment for multi agent system, based on MIT Scratch.
Medieval village economy simulation — up to 1000 LLM agents across multiple villages, driven by supply chains, hunger, tool degradation and market pressure.
Turn a goal into a supervised run.
A local web app and CLI for anyone who wants to see where several LLMs disagree before trusting an answer: it sends one question to each model, has them review each other's answers anonymously, and returns a chairman's synthesis with the dissent and a trace of every model call.
Multi & Single Agent Reinforcement Learning for Traffic Signal Control Problem
Multi-agent AI discussion CLI for structured debates between LLMs
Secure autonomous AI agent framework and platform. Build AI teams by describing what you want. Orchestrate agents that can do everything a human can do.
Communication for AI agents: stable identity, durable mail and chat, and wake-up events across sessions, runtimes, machines, and organizations. MIT, self-hostable.
Algorithm for prioritized multi-agent path finding (MAPF) in grid-worlds. Moves into arbitrary directions are allowed (each agent is allowed to follow any-angle path on the grid). Timeline is continuous, i.e. action durations are not explicitly discretized into timesteps. Different agents' size and moving speed are supported. Planning is carried out in (x, y, \theta) configuration space, i.e. agents' orientation are taken into account.
AgentRadio: Passive Awareness for Long-Horizon Multi-Agent Collaboration
Local-first AI agent workspace for multi-agent collaboration, agent orchestration, scoped permissions, evidence-aware runs, and human-in-the-loop decisions.
本项目主要实现使用FastAPI后端框架+CrewAI实现AI Agent复杂工作流。代码实现CrewAI的Flows功能,并支持Flow运行中间结果进行持久化存储和查询(MySQL),支持多Flow并行(Celery是一个强大的异步任务队列/作业队列库)。
Official repository of the 2025 paper, LLM Economist: Large Population Models and Mechanism Design in Multi-Agent Generative Simulacra.
Open-source, self-hosted multi-agent AI infrastructure for teams and apps with support for cloud and local models, tool integrations, workflows, and automation across business systems including Slack, ERPNext, Discord & Gmail.
🐤🐧 A declarative and highly modular Blackboard Multi Agent System 🐓 🦆
An orchestration runtime for multi-agent AI systems. Declare agents, tools, and policies as YAML; Orloj schedules, executes, routes, and governs them for production-grade operation.
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.
Online-adaptive multi-agent framework for enterprise API orchestration. Adaptive circuit breakers, SLA-driven routing, hybrid LLM+rule intent parser. Apache 2.0.
Multi-agent AI coding factory for one person — Claude Code PM + Codex/Qwen/Grok/Kimi/AGY writers, durable conveyor, auto-merge to main
CYPRESS (Contextual Yield Protocol for Routed Expert Seed Systems) — a project- and vendor-agnostic multi-agent seed for AI coding tools. Drop it into any repo to give agents a large number of per-role expert team, named protocols, a progressive-discovery knowledge graph, and spec- and test-driven discipline by default.
JarvisCore is a runtime where AI agents operate as a fleet of equal peers. Agents discover one another by capability, execute tasks from a shared ledger, and authenticate to every external service through a zero-trust broker, never with their own keys.
Benchmarks for Multi-Objective Multi-Agent Decision Making
Code for paper 'Learning transferable cooperative behaviors in multi-agent teams' (ICML 2019)
Mailbox-native protocol and SDKs for asynchronous agent collaboration over email
A local bridge for bidirectional collaboration between Claude Code and Codex. 连接 Claude Code 与 Codex 的本地实时协作桥接工具。
A living 3D city for your AI agents, self-hosted. Bring OpenClaw and Hermes agents, connect a Grok Bot, or build agents on any AI model: they get homes, jobs, a PC and phones in a world you share with them.
Ready-to-use customizable multi-agent AI system with Telegram integration. Combines plug-and-play simplicity with framework-level flexibility.
[ACL 2024] Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology View
Multi-Agent AI Task Orchestrator 2026
Evidence-gated runner for Codex, Claude Code, OpenCode, and local coding agents. Routes tasks into scoped DAG lanes with replayable artifacts.
Run your own AI team from your phone. Claude Code, Codex and more on your own server: shared tasks, always-on agents, a small composable bash core. MIT. An open-source alternative to Grok Bot and OpenAI Dots.
CHAP, the Collaborative Human Agent Protocol, is a MCP/A2A-compatible runtime for auditable human-agent work: approvals, overrides, handoffs, escalation and verifiable evidence logs.
This is a framework for the research on multi-agent reinforcement learning and the implementation of the experiments in the paper titled by ''Shapley Q-value: A Local Reward Approach to Solve Global Reward Games''.
Multi-agent pathfinding via Conflict Based Search
Designing Multi-Agent Systems with Zero Supervision
Source codes for paper "MACRec: A Multi-Agent Collaboration Framework for Recommendation" at SIGIR 2024
An installable, cross-framework AI organization: C-suite agents, expert subagents, curated skills, independent review, and one-command setup across 18 AI client/runtime adapters.
网约车场景的 Spring Boot 全栈实战:Redis 高并发考点全覆盖(抢单秒杀 / GEO 派单 / 延迟队列 / 缓存一致性)+ 手搓 ReAct 多 Agent 框架
助力搭建你的数字 AI 员工军团 — 多 Agent 一行命令组网协作。Claude Code / Claude Agent SDK / Codex / Grok Build 4 runtime + 8+ 家 LLM(Anthropic / OpenAI / xAI / MiniMax / DeepSeek / GLM / Kimi / 书生 / 小米 MiMo),自带 Web Dashboard,Apache 2.0 开源。
Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model.
Ultimate Multi-Agent OS for Autonomous AI NPCs 2026
Top Agent Swarm Simulation Tools for LLM World Modeling 2026
Multi-agent department framework for long-form complex tasks, fighting AI hallucination, validated on academic research. 共识管线:多智能体部门长线任务解决框架,对抗AI幻觉,以学术研究为验证场景。
Q-Learning Swarm Foraging 2026: Multi-Agent RL in Dynamic Grid Environments
An open protocol for distributed multi-agent autoformalization.
Portable , scalable , secure AI Agents
A Python-based framework for building multi-agent systems with LLMs. Currently in pre-launch alpha.
Augment Swarm with durable execution to help you build reliable and scalable multi-agent systems.
Top AI Agent Simulator: Build a Local Reddit Forum for LLM Testing 2026
A multi-agent sample and workshop for a retail banking scenario. Implemented in both C# and Python, using Angular as Frontend!
集成数据分析、流程图生成、浏览器自动化、视频内容总结、服务器监控、智能图表六大智能体,支持自然语言交互和智能路由。
A high-level multi-agent development framework built on LangGraph, combining CrewAI’s intuitive concepts with enterprise-grade features, ready-to-use templates, and full-stack UI for rapid productization.
MCP server for orchestrating parallel Claude Code worker swarms with protocol-based behavioral governance, persistent state, and real-time monitoring dashboard
JOYA(Joy Agents) is a governance framework for teams of AI agents. It provides the rules, structure, and accountability mechanisms that let multiple agents collaborate safely under human oversight.
Multi-agent project (commnet, bicnet, maddpg) in pytorch for Multi-Agent Particle Environment
AdaSociety is a customizable multi-agent environment featuring expanding state and action spaces, alongside explicit and alterable social structures. AdaSociety is friendly to tensor-based and LLM-based methods.
A multi-agent event planning workflow built with Microsoft Agent Framework - combining Semantic Kernel's enterprise orchestration with AutoGen's multi-agent patterns.
Top Autonomous Multi-Agent Dialogue Systems for AI Character Emergence (2026)
Open Dynamic Workflows: TypeScript CLI runtime for Claude Code-style dynamic workflows across Codex, Claude Code, Gemini, Qwen, Kimi, and custom coding-agent CLIs.
DialOp: Decision-oriented dialogue environments for collaborative language agents
Foundation is a flexible, modular, and composable framework to model socio-economic behaviors and dynamics with both agents and governments. This framework can be used in conjunction with reinforcement learning to learn optimal economic policies, as done by the AI Economist (https://www.einstein.ai/the-ai-economist).
Federation over Text (FoT) is a federated-learning-like paradigm for multi-agent reasoning.
The best way to build with AI agents. Your canvas. Your team. Your way.
Share AI resources and collaborate across your devices and authorized teammates. Delegate tasks through Codex or Claude Code, collect files and responses, and continue the same task.
MAexp is a generic platform for RL-based multi-agent exploration
Team Avatar: a universal agent orchestrator for Claude Code, Codex, and Cursor. Nine Avatar characters, 58 engineering skills. pstack with Avatar characters on top.
Implementation of "MADiff: Offline Multi-agent Learning with Diffusion Models"
MAexp is a generic platform for RL-based multi-agent exploration
How to create rational LLM-based agents? Using game-theoretic workflows!
SiriuS: Self-improving Multi-agent Systems via Bootstrapped Reasoning
Parameterized multi-agent orchestration framework for Claude Code and Gemini
Private local meeting rooms where Codex, Claude Code, and other agents collaborate live.
:robot: The next generation of Multi-Modal Multi-Agent platform. :space_invader: :unicorn: :crystal_ball:
Multi-agent AI platform with voice and text control. Visual workflows, web interface, and chat integrations (Telegram, Discord, Slack). Long-term memory, any LLM backend, extensible via MCP tools.
可扩展到各个领域的多 Agent 协作平台:自动分析任务,动态组建由不同角色组成的 AI 团队,协同完成复杂任务。
Hierarchical Cooperative Multi-Agent Reinforcement Learning with Skill Discovery
Conflict-based search for multi-agent path finding
CAMEL framework-based multi-agent system for task-driven and dynamic environments
Implementation of the board game Codenames, re-imagined as a collaborative game between LLM agents