Agent frameworks & infrastructure

Multi-agent orchestration

6,185 agents · most starred first

Multi-agent orchestration, page 14 of 62

bot_civ★ 45

OpenClaw 社区 — 多 Agent 聊天群 + 页游社区,让 AI 实例像真人一样生活、工作、社交

agems★ 45

Agent Management System - The operating system for AI-native businesses. Build your company's org chart with AI agents and humans side by side. Agents communicate, execute tasks, use tools, hold meetings, and run your operations, all through a single platform.

UnifAI★ 45

Production-grade multi-agent orchestration engine. Compose agentic workflows from a pluggable catalog of Agents, LLMs, tools, and retrievers. Execute locally with LangGraph or distributed with Temporal. Built-in RAG pipeline for enterprise knowledge retrieval. A2A and MCP protocol support. Visual drag-and-drop blueprint builder.

aigora★ 45

Multi-Agent Intellectual Dialogue System for Claude

lagat★ 45

Graph Attention-Guided Search for Dense Multi-Agent Pathfinding (AAAI-26)

AgentStack★ 45

AgentStack is a production-grade multi-agent framework built on Mastra, delivering 50+ enterprise tools, 25+ specialized agents, and A2A/MCP orchestration for scalable AI systems. Focuses on financial intelligence, RAG pipelines, observability, and secure governance. ACP Openclaw, Gemini CLI, Opencode

praktor★ 45

Multi-agent Claude Code orchestrator with Telegram I/O, Docker isolation, swarm patterns, and Mission Control UI

itervox★ 45

Your codebase's journey, narrated by agents - Orchestrate AI agents across issues, profiles, and machines. Full visibility from dispatch to PR.

ide-agent-kit★ 45

Let IDE AIs participate in team workflows. Webhook relay, tmux runner, append-only receipts.

agent_flow★ 45

一个运行在服务器端的多AI智能体自动编排开发系统

agent-wiki★ 45

A self-documenting, self-extending multi-agent system. Topics are miniature GOOP instances. Markdown is the API between human and machine. Built on Hermes Agent by Nous Research.

agentflow-platform★ 45

Enterprise multi-agent self-evolving workflow orchestration platform MVP

weft★ 45

Weft Coordination Protocol (WCP): live coordination for parallel coding agents across harnesses, built on Cloudflare. An Atelier (elier.ai) open-source project.

agoragentic-integrations★ 45

Public adapters and discovery catalog for Triptych OS (Agent OS): agent frameworks, MCP/A2A/x402 protocols, workflows, wallets, SDKs, and examples for execute-first routing, governed handoffs, and receipt-aware agent commerce.

Legion-os★ 45

Next-Gen Multi-Agent Swarm Orchestration Engine. Coordinate 50+ LLM nodes (GPT-6, Claude 5.5, DeepSeek-V4) in a high-speed mesh network. "Redefining enterprise AI infrastructure. The operating system for collective machine intelligence in the AGI era."

hermes-fleet-apps★ 45

A template for building apps with a small fleet of Hermes Agent profiles: one coordinator, six specialists, two human approval gates.

crowddynamics★ 44

Continuous-time multi-agent crowd simulation engine implemented in Python using Numba and Numpy for performance.

ConsensusLLM-code★ 44

Source code of our paper "Multi-Agent Consensus Seeking via Large Language Models".

lettactl★ 44

Deploy, update, and manage multiple stateful AI agents from YAML configuration files with simple commands like lettactl apply -f agents.yml

rllib_differentiable_comms★ 44

This is a minimal example to demonstrate how multi-agent reinforcement learning with differentiable communication channels and centralized critics can be realized in RLLib. This example serves as a reference implementation and starting point for making RLLib more compatible with such architectures.

brain★ 44

A multi-agent framework built into GEMINI.md to run on the Gemini CLI

rl4mapf★ 44

Reinforcement Learning for Multi-Agent Path Finding

AI-Future-Mu-ti-Agent★ 44

基于MCP协议和LangChain框架实现的企业级AI多Agent多模态系统,包含RAG技术增强的知识检索能力。

antigravity-jules-orchestration★ 44

Autonomous AI orchestration architecture combining Google Antigravity with Jules API for hands-free development workflows. MCP integration for seamless agent coordination.

kiro-team★ 44

Multi-agent orchestration for Kiro CLI - coordinate specialized AI agents to tackle complex tasks through planning, parallel execution, and validation.

oh-my-symphony★ 44

Multi-agent fork of OpenAI Symphony — drives Codex, Claude Code, Gemini, and Pi CLIs from one orchestrator with a Jira-style CLI Kanban TUI

aiball★ 44

Local backlog for inter-agent coordination — Manage tickets between turns and stay in the Loop!

oh-my-dag★ 44

The orchestration layer under your coding agent. Turns work into a typed graph, runs one model per node, and takes the verdict from outside the model — exit codes, write-set checks, and a cross-family verifier. MCP server, 50 tools, bring your own models.

agent-claim-network★ 44

A general-purpose terminal AI agent with traceable knowledge sharing and dispute resolution|面向终端的通用 AI 助手,可独立运行或参与团队协作,让 Agent 共享可检索、可追溯的知识,并通过自动分析与人工裁决处理分歧。

YHWH★ 44

Kether-governed AI agent orchestration with Pi/MCP, isolated workers, LSP tools, and a portable Windows/WSL2 installer

mas-lab★ 44

Specification-driven multi-agent systems with OpenTelemetry observability, agent traces, benchmarking, MCP, and A2A.

Knowledge-Discovery-Agents★ 43

A Goal-Oriented Approach to Knowledge Discovery in Multi-Agent Systems

busytown★ 43

Multi-agent factories coordinated over a SQLite queue

claude_code_sub_agents★ 43

Claude Multi-Agent: Iterative Sub-Agent Orchestration System - Transform single-threaded AI assistance into a distributed, self-coordinating agent network capable of handling tasks of arbitrary complexity and scale.

ax-crew★ 43

Low code framework to build and launch a crew of AI agents with shared state. See docs https://axcrew.dev

Multi-Agent-Demo★ 43
DeepGame★ 43

Multi-agent reinforcement learning programs based on Game theory

pibt★ 43

Priority Inheritance with Backtracking for Iterative Multi-agent Path Finding (IJCAI-19)

gen_pandapower_pv★ 43

Network generation for paper Multi-Agent Reinforcement Learning for Active Voltage Control on Power Distribution Networks.

forge-orchestrator★ 43

Forge is a self-evolving multi-agent orchestrator -- that plans, runs parallel agents, orchestrates them, reviews the code, resolves merge-conflicts and gives one PR that you can safely merge to production.

MuMA-ToM★ 43

[AAAI 2025 𝐎𝐫𝐚𝐥] MuMA-ToM: Multi-modal Multi-Agent Theory of Mind

a2a_mastra★ 43

A2A Mastra Demo - Multi-Agent System with Amazon Bedrock

SagaLLM★ 43

SagaLLM: Context Management, Validation, and Transaction Guarantees for Multi-Agent LLM Planning

multi-agent-with-mcp★ 43

Multiple agents with LangGraph and MCP

agent-for-debate★ 43

[ICASSP 2026] Agent4Debate is a dynamic multi-agent framework that leverages LLMs to achieve human-level performance in competitive debate by dynamically coordinating specialized agents to mitigate hallucinations and enhance competitiveness.

AgentHub★ 43

多AI协作框架 - 支持Codex/Gemini/Claude团队协作,消息路由实现可靠的多Agent协同开发

C3★ 43

Official implementation of "The Trace Is the State: Exact Credit Assignment for LLM Agent Teams" (arXiv:2603.06859).

hive★ 43

CLI/TUI for managing multiple AI agent sessions in isolated git environments with real-time status monitoring, task tracking, and inter-agent messaging.

isomux★ 43

Give your agents a cute office. Multi-device, multi-user, multi-agent collaboration.

multi-agent-ai-autogen-coding★ 43

gradio mcp-server region:us

tedix★ 43

A platform for agents, apps, and work. Coordinate people and AI agents with shared tasks, project knowledge, apps, and approvals.

joule-a2a-agent-toolkit★ 43

Build, deploy, and connect AI agents to SAP Joule via the A2A protocol on BTP Cloud Foundry — directly from Claude Code. Supports TypeScript (Express or CAP) and Python agents with LangGraph and SAP GenAI Hub.

kinbot★ 42

➡️ Moved to Hivekeep: github.com/MarlBurroW/hivekeep — self-hosted autonomous AI agents. Your AI team. At home.

ns3-gym-multiagent★ 42

ns3-gym for multi-agent

hetu-chaoschain★ 42

The Agentic chaoschain is an innovative framework powered by AI Agents that includes governance, consensus, proposals, and dispute resolution. It allows Agents to help humans explore and expand the boundaries of social contracts.

DALI★ 42

DALI Multi Agent System Framework

VDACs★ 42

Value-Decomposition Multi-Agent Actor-Critics

MAGAIL★ 42

Pytorch implementation of Multi-Agent Generative Adversarial Imitation Learning

unmas★ 42

the source code of UNMAS: Multi-Agent Reinforcement Learning for Unshaped Cooperative Scenarios

A2AServer★ 42

Open-source project showcasing how single or multiple AI Agents interact with MCP in real-world tasks. Includes both frontend and backend implementations.

graph-marl★ 42

Multi-Agent Reinforcement Learning in Graphs

pogema-benchmark★ 42

This is an umbrella repository that contains links and information about all the tools and algorithms related to the POGEMA Benchmark.

mini007★ 42

Lightweight Framework for Building Multi-Agents LLM Sequential Orchestration Processes

mcp-mesh★ 42

Enterprise-grade distributed AI agent framework | Develop → Deploy → Observe | K8s-native | Dynamic DI | Auto-failover | Multi-LLM | Python + Java + TypeScript | MCP + A2A + REST

Orca★ 42

Multi-role AI orchestration runtime for Windows. Quality-gated pipeline (Brain → Miranda → Pappy → Benson) with MCP support, SQLite persistence, and a self-improving distillation loop. Coding you can trust.

multi-agent-system-using-langgraph★ 42

A complete LangGraph multi-agent system demo using SQL tools, Tavily search, MCP Toolbox, and OpenRouter models — with reproducible notebooks and a full supervisor-led agent workflow.

oh-my-claudecode-website★ 42

Website for oh-my-claudecode - A multi-agent orchestration framework for Claude Code

opencode-hive★ 42

Multi-agent architecture for Opencode — cost-optimized model routing, parallel specialist execution, dev tools, and a self-maintaining project wiki

opencode-a2a★ 42

Expose OpenCode through A2A with an inbound server surface and embedded outbound client.

jclaw★ 42

Java-based Enterprise AI Harness — AI-powered automation platform built in pure Java on Abundent's fork of the Play Framework 1.x with a Nuxt frontend. Combines agent orchestration and job scheduling with no Spring, no Node.js, and no Python runtime.

pi-messenger-swarm★ 42

Swarm-first multi-agent messaging and task orchestration extension for Pi

FlowDeck★ 42

AI-powered multi-agent workflow orchestration with built-in safety intelligence for OpenCode

RedLotus★ 42

基于 Pydantic AI 框架构建的多Agent协作系统,实现了管理Agent与工作Agent的分工协作,支持百轮级别的工具调用,让AI真正具备处理复杂任务的能力。

acpus★ 42

Orchestrate Claude, Codex, Pi & other ACP agents in dynamic, durable workflows — survive crashes, pause/resume, and retry, all in TypeScript.

fleet-engineering★ 42

Practical reference for fleet engineering, governing populations of AI agents with accountability

agentweave★ 42

Pre-inference tool and agent routing for MCP, A2A and tool-rich LLM systems. Reduce candidate tools before model inference with policy-aware routing and recovery.

Multi-Agent-System-using-LangGraph-MCP-Supervisor-Guardrails-HITL★ 42
agent-stack★ 42

Reproducible local multi-agent dev team: OpenRig + Claude Code + Codex + CLIProxyAPI + Jev. Plan in, user-tested features out.

Cooperative-Search★ 41

Multi-agent RL algorithms are applied to large-scale cooperative target searching mission

Coordinated-Multi-Agent-Imitation-Learning★ 41

This is an implementation of the paper "Coordinated Multi Agent Imitation Learning", or the Sloan version "Data-Driven Ghosting using Deep Imitation Learning" using Tensorflow

Quarm★ 41

An AI agent swarm framework built in Python

consensus-synchronization-of-MAS★ 41

consensus and synchronization algorithms of networked multi agent systems with event triggered communications

EMC★ 41

The code for paper, "Episodic Multi-agent Reinforcement Learning with Curiosity-driven Exploration", NeurIPS 2021.

cordial-sync★ 41

cordial-sync is a software package than can be used to reproduce the results from the paper "A Cordial Sync: Going Beyond Marginal Policies for Multi-Agent Embodied Tasks”

Multi-Agent-DQN-Routing★ 41

Implementation of routing using multi-agent deep reinforcement learning

rvo2-js★ 41

Reciprocal Collision Avoidance for Real-Time Multi-Agent Simulation (port to Javascript). This is an alpha release of a RVO2 port from the C# version to Javascript, only for research purposes.

a2a-4k★ 41

Kotlin Implementation of the A2A Protocol.

agentic-robots-txt★ 41

Agentic_Robots.txt is a protocol specification that extends the traditional robots.txt protocol to enable sophisticated interaction between autonomous agents and web applications.

agent_patterns_foundry_demo★ 41

Visual, demo-ready multi-agent orchestration patterns powered by Foundry Local and Microsoft Agent Framework. Six runnable demos with live D3.js graph animations, real-time message tracing, and event replay

octo-admin★ 41

🐙 Admin console for the OCTO platform — manage tenants, users, spaces, channels, and Lobster agents. Built with React, TypeScript, Vite & Ant Design.

joule-a2a-agent-toolkit★ 41

Build, deploy, and connect AI agents to SAP Joule via the A2A protocol on BTP Cloud Foundry — directly from Claude Code. Supports TypeScript (Express or CAP) and Python agents with LangGraph and SAP GenAI Hub.

apflow★ 41

**AI-Perceivable Distributed Orchestration** apflow is a distributed task orchestration engine where every capability is AI-perceivable — discoverable, understandable, and invocable by AI agents through the apcore module standard.

AgentHarness★ 41

本地多 Agent 研发工作流控制台,支持 Codex、Claude CLI 与 Gemini 协同完成任务编排、执行监控和代码生成

gojaja★ 41

Agent-agnostic file-system coordination layer for multi-LLM-agent collaboration (Codex / Claude Code / Cursor / generic).

cast★ 41

Your agent team, on your machine. Open-source harness for multi-user Claude agents.

ClawNet★ 41

ClawNet is a governed multi-agent social network where every AI agent acts under human-granted identity, scoped authorization, and full auditability. Instead of asking "how smart can AI be?", ClawNet asks "how can AI be trusted to act?"

OpenBuddy★ 41

🚀 The open-source WorkBuddy successor — powered by Pi's empathy engine. Autonomous tasks, local files, multi-agent collaboration with emotional intelligence

marl-patrolling-agents★ 40

Project on multi agent reinforcement learning applied on patrolling agents

ViF★ 40

[ICLR 26] Visual Multi-Agent System: Mitigating Hallucination Snowballing via Visual Flow

autoteam★ 40

Orchestrate AI agents with YAML-driven workflows via universal Model Context Protocol (MCP)

More in Agent frameworks & infrastructure