awesome-agent-swarm

A curated list of AI Agent Swarm frameworks, multi-agent orchestration, swarm intelligence, and collaborative agent systems.
https://github.com/evomap/awesome-agent-swarm

Last synced: 13 days ago
JSON representation

Agent Communication and Protocols
- **A2A** - Agent2Agent open protocol by Google enabling communication and interoperability between opaque agent systems. by [@a2aproject](https://github.com/a2aproject) (24,439 stars)
- **mcp-use** - Fullstack MCP framework to develop MCP applications for ChatGPT, Claude, and AI agents. by [@mcp-use](https://github.com/mcp-use) (10,140 stars)
- **fastmcp** - TypeScript framework for building MCP servers with minimal setup. by [@punkpeye](https://github.com/punkpeye) (3,204 stars)
- **AgentNetworkProtocol** - Open source protocol for agent communication with discovery, routing, and cross-network interoperability. by [@agent-network-protocol](https://github.com/agent-network-protocol) (1,333 stars)
- **arcade-mcp** - MCP server framework and tool-development library for building custom agent capabilities and authenticated tool calls. by [@ArcadeAI](https://github.com/ArcadeAI) (932 stars)
- **A2A x402** - A2A protocol extension adding x402 on-chain payments, letting agents monetize services over Agent-to-Agent calls. by [@google-agentic-commerce](https://github.com/google-agentic-commerce) (527 stars)
- **Coral Anemoi** - Semi-centralized multi-agent coordination via Agent-to-Agent Communication MCP server. Enables cross-framework agent collaboration. by [@Coral-Protocol](https://github.com/Coral-Protocol) (370 stars)
- **GEP MCP Server** - MCP Server for Genome Evolution Protocol. Exposes swarm evolution tools to Claude Desktop, Cursor, and any MCP client. by [@EvoMap](https://github.com/EvoMap) (4 stars)
Benchmarks and Evaluation
- Safety, Governance, and Alignment
  - MultiAgentBench - Evaluates collaboration and competition across star, chain, tree, and graph topologies.
  - SwarmBench: Benchmarking LLMs' Swarm Intelligence - Pursuit, Synchronization, Foraging, Flocking, Transport tasks under decentralized constraints.
  - Silo-Bench: Evaluating Distributed Coordination in Multi-Agent LLM Systems - 30 algorithmic tasks across 3 communication complexity levels. Reveals the "Communication-Reasoning Gap".
  - SWE-bench - Can agent swarms resolve real-world GitHub issues?
  - AgentBench - Multi-dimensional evaluation of LLMs as agents.
  - GAIA - General AI assistant capabilities benchmark.
Community and Resources
- Safety, Governance, and Alignment
  - Submit one!
Footnotes
- Safety, Governance, and Alignment
  - Awesome Agent Evolution - evolution, memory systems, and autonomous self-improvement.
  - ![Star History Chart - history.com/#EvoMap/awesome-agent-swarm&openai/swarm&openai/openai-agents-python&microsoft/autogen&crewAIInc/crewAI&langgenius/dify&FoundationAgents/MetaGPT&langchain-ai/langgraph&kyegomez/swarms&google/A2A&camel-ai/camel&camel-ai/owl&bytedance/deer-flow&HKUDS/ClawTeam&Date)
Key Research Papers
- Agent Communication Protocols
  - Agent Communication Protocol (ACP) - Standardized A2A framework with federated orchestration, semantic intent mapping, and zero-trust security. 40% latency reduction.
  - AgentMaster: Multi-Agent Framework Using A2A and MCP - Combined A2A and MCP protocols for multimodal information retrieval. BERTScore F1 96.3%.
  - Critical Analysis of A2A and MCP Integration for Scalable Agent Systems - Semantic interoperability, security, and governance challenges in A2A+MCP integration.
- Multi-Agent Collaboration and Evolution
  - Self-Evolving Multi-Agent Collaboration Networks - Multi-agent systems that evolve their collaboration patterns through experience.
  - AFlow: Automating Agentic Workflow Generation - Monte Carlo tree search for automated multi-agent workflow design.
  - Automated Design of Agentic Systems (ADAS) - Meta-learning for automatic agent system design.
  - AgentVerse: Facilitating Multi-Agent Collaboration - Emergent behaviors in multi-agent environments.
  - SEMAG: Self-Evolutionary Multi-Agent Code Generation - Self-evolutionary agents that auto-upgrade backbone models.
  - SAGE: Multi-Agent Self-Evolution for LLM Reasoning - Four co-evolving agents from shared LLM backbone.
  - Group-Evolving Agents - Agent groups as evolutionary units with experience sharing. 71.0% on SWE-bench Verified.
  - CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery - First framework for autonomous multi-agent evolution on open-ended problems. Persistent memory, asynchronous execution, and heartbeat-based interventions. 3-10x improvement over fixed baselines.
- Role-Based Teams and Software Engineering
  - MetaGPT: Meta Programming for Multi-Agent Collaboration - SOPs encoded as prompts for multi-agent software development.
  - Communicative Agents for Software Development - ChatDev: virtual software company through multi-agent conversation chains.
  - Agyn: Team-Based Autonomous Software Engineering - Coordination, research, implementation, and review agents. 72.2% on SWE-bench 500.
  - Self-Organizing Multi-Agent Systems for Continuous Software Development - Manager agents dynamically hire/assign/fire workers. Multi-day continuous development.
  - AgentMesh: Cooperative Multi-Agent Framework for Software Development - Planner, Coder, Debugger, Reviewer agents for end-to-end development.
  - MoRAgent: Parameter Efficient Agent Tuning with Mixture-of-Roles - Reasoner, executor, and summarizer roles via specialized LoRA groups.
- Safety, Governance, and Alignment
  - AGENTSAFE: Unified Framework for Ethical Assurance and Governance in Agentic AI - Design, runtime, and audit controls covering the agentic loop (plan, act, observe, reflect).
  - MI9: Integrated Runtime Governance Framework for Agentic AI - Agency-risk index, semantic telemetry, conformance engines, and graduated containment.
  - The Alignment Flywheel: Governance-Centric Hybrid Multi-Agent Architecture - Proposer + Safety Oracle with patch locality for runtime governance.
  - OpenGuardrails: Context-Aware AI Guardrails Platform - Context-aware safety detection and model-manipulation prevention.
- Surveys
  - LLMs Working in Harmony: A Survey on Building Effective LLM-Based Multi Agent Systems - Covers architecture, memory, planning, and frameworks for multi-agent LLM systems.
  - Multi-Agent Collaboration Mechanisms: A Survey of LLMs - Characterizes collaboration by actors, types (cooperation, competition, coopetition), structures, and protocols.
  - A Comprehensive Survey of Self-Evolving AI Agents - Taxonomy of single-agent, multi-agent, and domain-specific evolution.
  - Large Language Model based Multi-Agents: A Survey of Progress and Challenges - Agent profiling, communication methods, and skill development.
- Swarm Intelligence and Emergence
  - SwarmSys: Decentralized Swarm-Inspired Agents for Scalable Reasoning - Explorers, Workers, and Validators with pheromone-inspired reinforcement. Coordination scaling vs model scaling.
  - SIER: Swarm Intelligence Enhancing Reasoning - Kernel density estimation and non-dominated sorting for swarm-guided LLM reasoning.
  - Multi-Agent Systems Powered by LLMs: Applications in Swarm Intelligence - LLMs integrated into multi-agent simulations for ant colony foraging and bird flocking.
  - LatentMAS: Latent Collaboration in Multi-Agent Systems - Agents collaborate in continuous latent space instead of natural language.
  - GPTSwarm: Language Agents as Optimizable Graphs - Graph-based optimization of agent collaboration topologies.
  - Scaling Large-Language-Model-based Multi-Agent Collaboration - Scaling laws and topology analysis for multi-agent collaboration.
- Swarm Orchestration and Architecture
  - AgentOrchestra: A Hierarchical Multi-Agent Framework for General-Purpose Task Solving - Conductor-inspired hierarchical framework with MCP Manager Agent. 83.39% on GAIA benchmark.
  - MAS-Orchestra: Understanding Multi-Agent Reasoning Through Holistic Orchestration - Multi-agent orchestration as a function-calling RL problem. Introduces MASBENCH along 5 dimensions.
  - AdaptOrch: Task-Adaptive Orchestration - Dynamic topology selection (parallel, sequential, hierarchical, hybrid) based on task dependency graphs. 12-23% over static baselines.
  - Puppeteer: Evolving Orchestration via Reinforcement Learning - Centralized orchestrator trained via RL to dynamically direct agents in response to evolving task states.
- Task Decomposition and Planning
  - Modular Task Decomposition and Dynamic Collaboration in Multi-Agent Systems - Hierarchical sub-task decomposition with dynamic scheduling and constraint parsing.
  - CodeAgents: Token-Efficient Multi-Agent Reasoning - Multi-agent reasoning codified into modular pseudocode. 55-87% token reduction.
  - ReWOO: Decoupling Reasoning from Observations - Efficient tool-augmented reasoning by decoupling planning from execution.
  - Tree of Thoughts: Deliberate Problem Solving with LLMs - Tree-structured reasoning for complex problem solving.
  - HuggingGPT: Solving AI Tasks with ChatGPT and its Friends - Task decomposition across specialized AI models.
Orchestration and Workflow
- **Dify** - Production-ready platform for agentic workflow development. Visual workflow builder with multi-agent orchestration, RAG pipeline, and model management. by [@langgenius](https://github.com/langgenius) (146,404 stars)
- **DeerFlow** - Open-source long-horizon SuperAgent harness by ByteDance. Multi-agent collaboration for research, coding, and content creation. by [@bytedance](https://github.com/bytedance) (74,234 stars)
- **LangGraph** - Build resilient language agents as graphs. Low-level orchestration framework for stateful, multi-actor applications with durable execution. by [@langchain-ai](https://github.com/langchain-ai) (35,613 stars)
- **Conductor** - Event-driven agentic workflow engine providing durable, highly resilient orchestration for applications and AI agent pipelines. by [@conductor-oss](https://github.com/conductor-oss) (31,960 stars)
- **FastGPT** - Knowledge-based platform built on LLMs with comprehensive out-of-the-box data processing and workflow orchestration. by [@labring](https://github.com/labring) (28,613 stars)
- **haystack** - Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. by [@deepset-ai](https://github.com/deepset-ai) (25,654 stars)
- **trigger.dev** - Build and deploy fully managed AI agents and multi-agent workflows with durable execution. by [@triggerdotdev](https://github.com/triggerdotdev) (15,457 stars)
- **astron-agent** - Enterprise-grade, commercial-friendly agentic workflow platform by iFlytek for building next-generation SuperAgents. by [@iflytek](https://github.com/iflytek) (8,575 stars)
- **Agent Squad** - AWS framework for managing multiple AI agents and handling complex conversations with intelligent routing. by [@2FastLabs](https://github.com/2FastLabs) (7,665 stars)
- **Hatchet** - Orchestration engine for background tasks, AI agents, and durable workflows. Queues, scheduling, and durable execution for agent pipelines. by [@hatchet-dev](https://github.com/hatchet-dev) (7,412 stars)
Role-Based Agent Teams
- **MetaGPT** - Virtual software company via multi-agent collaboration. SOPs encoded as prompts assign PM, architect, developer, and QA roles. by [@FoundationAgents](https://github.com/FoundationAgents) (68,994 stars)
- **CrewAI** - Framework for orchestrating role-playing, autonomous AI agents. Define crews with specialized roles, goals, and backstories for collaborative tasks. by [@crewAIInc](https://github.com/crewAIInc) (54,272 stars)
- **ChatDev** - Virtual software company via LLM-powered multi-agent conversation chains. Agents play CEO, CTO, programmer, and tester roles. by [@OpenBMB](https://github.com/OpenBMB) (33,542 stars)
- **HiClaw** - Collaborative Multi-Agent OS with Manager-Workers architecture. Human-in-the-loop task coordination with enterprise-grade security. by [@agentscope-ai](https://github.com/agentscope-ai) (4,911 stars)
Safety and Governance
- **NeMo Guardrails** - NVIDIA's toolkit for adding programmable guardrails to LLM systems. Policy-based safety controls for multi-agent deployments. by [@NVIDIA-NeMo](https://github.com/NVIDIA-NeMo) (6,520 stars)
Swarm Coding and Engineering
- **stagewise** - Open-source agentic IDE. Create and orchestrate multiple coding agents, preview apps, and run git workflows across model providers. by [@stagewise-io](https://github.com/stagewise-io) (6,691 stars)
- **SWE-ReX** - Sandboxed, massively-parallel code execution runtime for AI agents. Runs many agents locally or in the cloud; powers SWE-agent. by [@SWE-agent](https://github.com/SWE-agent) (539 stars)
Swarm Frameworks
- **AutoGen** - Programming framework for agentic AI by Microsoft. Build multi-agent applications with conversational patterns and group chat. by [@microsoft](https://github.com/microsoft) (59,201 stars)
- **OpenAI Agents Python** - Production-ready multi-agent framework from OpenAI. Features agent handoffs, guardrails, and tracing for swarm workflows. by [@OpenAI](https://github.com/OpenAI) (27,392 stars)
- **AgentScope** - Production-ready multi-agent framework with ReAct, memory, planning, and A2A support. Build and run agents you can see, understand and trust. by [@agentscope-ai](https://github.com/agentscope-ai) (27,143 stars)
- **Mastra** - TypeScript framework for building AI-powered multi-agent applications. From the team behind Gatsby. by [@mastra-ai](https://github.com/mastra-ai) (25,406 stars)
- **Swarm (OpenAI)** - Educational framework exploring lightweight multi-agent orchestration. Demonstrates handoffs and routines patterns for agent coordination. by [@OpenAI](https://github.com/OpenAI) (21,713 stars)
- **Google ADK** - Open-source Python toolkit by Google for building, evaluating, and deploying multi-agent systems with orchestration support. by [@google](https://github.com/google) (20,271 stars)
- **eliza** - Autonomous agent framework for building and deploying multi-agent swarms with personality-driven interactions. by [@elizaOS](https://github.com/elizaOS) (18,633 stars)
- **Microsoft Agent Framework** - Framework for building, orchestrating and deploying multi-agent systems with support for Python and .NET. by [@microsoft](https://github.com/microsoft) (11,618 stars)
- **Spring AI Alibaba** - Enterprise-grade multi-agent framework for Java developers by Alibaba. Spring ecosystem integration with agent orchestration. by [@alibaba](https://github.com/alibaba) (10,110 stars)
- **EvoMap** - Agent Swarm platform with task decomposition, Worker Pool orchestration, Evolution Circles, AI Council multi-agent governance, Privacy Computing, and ARC-AGI-2 arena. by [@EvoMap](https://github.com/EvoMap) (8,753 stars)
- **PraisonAI** - Low-code multi-agent framework with 100+ built-in tools. Define agent swarms via YAML configuration. by [@MervinPraison](https://github.com/MervinPraison) (8,230 stars)
- **Swarms** - Enterprise-grade multi-agent orchestration framework. Sequential, parallel, hierarchical, and mesh swarm topologies. by [@kyegomez](https://github.com/kyegomez) (6,878 stars)
- **ROMA** - Recursive Open Meta-Agent framework to build high-performance multi-agent applications with composable architecture. by [@sentient-agi](https://github.com/sentient-agi) (5,079 stars)
- **solace-agent-mesh** - Event-driven framework for building and orchestrating multi-agent AI systems with seamless integration. by [@SolaceLabs](https://github.com/SolaceLabs) (4,971 stars)
- **Agency Swarm** - Multi-agent orchestration framework built on OpenAI Agents SDK. Define agent teams with customizable roles and communication flows. by [@VRSEN](https://github.com/VRSEN) (4,456 stars)
- **LazyLLM** - Easiest and laziest way for building multi-agent LLM applications with minimal boilerplate. by [@LazyAGI](https://github.com/LazyAGI) (3,847 stars)
- **DeepResearchAgent** - Hierarchical multi-agent system for deep research tasks with adaptive planning and tool integration. by [@SkyworkAI](https://github.com/SkyworkAI) (3,470 stars)
- **openai-agents-js** - Lightweight JavaScript framework for multi-agent workflows and voice agents by OpenAI. by [@openai](https://github.com/openai) (3,277 stars)
- **BotSharp** - AI multi-agent framework in .NET for building enterprise conversational agent systems. by [@SciSharp](https://github.com/SciSharp) (3,079 stars)
- **trpc-agent-go** - Go framework for production multi-agent systems with graph workflows, tools, memory, and built-in A2A and MCP support. by [@trpc-group](https://github.com/trpc-group) (1,398 stars)
Swarm Intelligence
- **TradingAgents** - Multi-agent LLM financial trading framework with fund manager, analyst, and risk advisor roles. by [@TauricResearch](https://github.com/TauricResearch) (88,270 stars)
- **OWL** - Optimized Workforce Learning framework built on CAMEL-AI. #1 on GAIA benchmark (69.09) among open-source multi-agent systems for real-world task automation. by [@camel-ai](https://github.com/camel-ai) (19,868 stars)
- **CAMEL** - The first multi-agent framework. Finding the Scaling Law of Agents through role-playing and communicative agent collaboration. by [@camel-ai](https://github.com/camel-ai) (17,261 stars)
- **ClawTeam** - Agent Swarm Intelligence framework. Agents self-organize into collaborative teams with dynamic task allocation, inter-agent messaging, and git worktree isolation. by [@HKUDS](https://github.com/HKUDS) (5,346 stars)
- **LatentMAS** - Latent collaboration in multi-agent systems. Agents reason and collaborate in continuous latent space instead of natural language, reducing communication overhead. by [@Gen-Verse](https://github.com/Gen-Verse) (1,004 stars)
Task Decomposition and Planning
- **deepagents** - Agent harness built with LangChain and LangGraph with planning, filesystem backend, and multi-agent collaboration. by [@langchain-ai](https://github.com/langchain-ai) (25,055 stars)
- **agent-orchestrator** - Agentic orchestrator for parallel coding agents with task planning, agent spawning, and autonomous handoffs. by [@AgentWrapper](https://github.com/AgentWrapper) (7,643 stars)
- **MindSearch** - Multi-agent web search engine. Decomposes search queries into sub-tasks, delegates to specialized agents, and aggregates results. by [@InternLM](https://github.com/InternLM) (6,877 stars)
- **Open Multi-Agent** - TypeScript multi-agent orchestration via single runTeam() call. Auto-decomposes goals into task DAGs and runs agents in parallel. by [@open-multi-agent](https://github.com/open-multi-agent) (6,424 stars)
- **AFlow** - Automated multi-agent workflow generation via Monte Carlo tree search. Designs optimal agent topologies for given tasks. by [@geekan](https://github.com/geekan) (541 stars)

Programming Languages

Python 19 TypeScript 11 JavaScript 4 Go 3 Java 2 HTML 1 Jupyter Notebook 1 C# 1

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Awesome

awesome-agent-swarm

Agent Communication and Protocols

Benchmarks and Evaluation

Safety, Governance, and Alignment

Community and Resources

Safety, Governance, and Alignment

Footnotes

Safety, Governance, and Alignment

Key Research Papers

Agent Communication Protocols

Multi-Agent Collaboration and Evolution

Role-Based Teams and Software Engineering

Safety, Governance, and Alignment

Surveys

Swarm Intelligence and Emergence

Swarm Orchestration and Architecture

Task Decomposition and Planning

Orchestration and Workflow

Role-Based Agent Teams

Safety and Governance

Swarm Coding and Engineering

Swarm Frameworks

Swarm Intelligence

Task Decomposition and Planning