Agent frameworks & infrastructure
Multi-agent orchestration
6,185 agents · most starred first
Multi-agent orchestration, page 23 of 62
Spawn and manage sub-agents in visible tmux panes — the orchestrator that makes multi-agent boss mode work for pi coding agent.
Local-first desktop app: 16 AI agents from 8 LLM providers debate any topic — with live argument mapping, fact-check badges, and conflict detection.
AI-powered simulation and prediction engine with interactive dashboard. Run scenarios, get instant AI-generated insights, and visualize results with professional charts.
TeamBench: Evaluating Agent Coordination under Enforced Role Separation
Local-first AI workspace for creators — browser-based, privacy-first, multi-agent collaboration
Agent Task Markdown: a Markdown runbook format and CLI for organizing agents into repeatable, inspectable, and coordinated workflows. 一种 Markdown 格式拓展和 CLI工具,用于把多个 agent 组织成可复用、可检查、可协同的工作流。
🚀 企业级多智能体编排与 ChatBI 数据洞察平台。支持多 Agent 协作、RAG 知识增强、物理数据源直连与安全沙箱执行、Redis 长期记忆引擎,并集成多模态 EmbedChat 挂件与全方位 Token 消耗审计看板。
Desktop workspace for managing agent trees, shared context, and live multi-agent coding sessions.
Multiplayer admin and orchestration layer for Hermes agents.
sgMAP is an open-source general multi-agent platform for building, orchestrating and scaling AI agents, workflows and research pipelines. sgMAP 是一个开源的通用多智能体平台,用于构建、编排和扩展 AI 智能体、工作流程和研究流程。
Not a framework. A file format for agent-to-agent handoffs — provenance, parallel safety, and human visibility by design. Model-agnostic. Framework-agnostic. Open spec, MPL-2.0.
Stop letting your experience disappear between sessions. Build your own coding-agent ecosystem: keep roles, skills, and lessons in files you can review, refine, and reuse. Start with one agent or coordinate teams in real terminals. Open source, local-first, no telemetry.
Your team's decisions, briefed to every Claude Code session before it writes code.
Multi-model consensus debate via the filesystem. LLMs propose, peer-review, rebut, vote and synthesize a group-confirmed answer. CLI + MCP.
Universal collaboration protocol + CLI for multi-agent LLM teams — Claude Code, Reasonix, Codex, WorkBuddy, Cursor
🐟 在 MiroFish 核心引擎基础上扩展而来的探索者版本:圆桌推演 + 政策推演 + 因果推断 + 博弈论 + Tauri 桌面端 + Command Workflow(命令工作流)。Forked from 666ghj/MiroFish (AGPL v3)
Turn a raw idea into a plan you can build. A multi-agent brainstorming methodology — isolated critique, fact-checking, and a build handoff to Claude Code.
Multi-agent orchestration system for OpenCode
Self-evolving 10-agent stack for one-person companies. Cron + eval + reflexion + skill library + PR-gated evolver. pip install solo-founder-os · 513 tests · MIT.
Canvas as a runtime for shaping agent-driven systems — a GitHub Copilot Canvas extension (Multi-Agent Dev Canvas) with a worked scenario, blog posts, and a demo prompt.
streamlit region:us
Open-source mission control for running multiple Claude Code agents across projects - one dashboard, a scheduler, cross-session memory, and browser/phone access. Runs locally on your own Claude subscription (the CLI, not the API). MIT, local-first.
把多会话 AI 协作组织成可接班、可审核、低上下文的部门制团队 · Codex Skill
Multi-agent pack for the Kiro CLI — one orchestrator, eight specialists, zero lock-in
/party — the skill that lets your agent sessions talk to each other. Claude Code, Cursor, Codex or any other agent: an open-standard skill plus the CLI behind it, on one machine or across machines.
Interactive multi-agent collaboration, meetings, group chats, and task execution for DeepSeek Harness.
Social simulation experiment with the goal is to make AI personas feel like people: noticing unevenly, replying late, lurking, masking, forming private alliances, misreading silence, and creating emergent social drama.
Your AI subscriptions, working as one team. Orchestrate Claude Code, Codex, Grok, Gemini & Antigravity CLIs as labeled tmux teammates, with quota-aware routing to whichever subscription has budget. Single-file Python, official vendor CLIs only.
Multi-agent demo with Google ADK on Vertex AI Agent Engine — every LLM, A2A, and MCP call routed through Apigee X for API-key auth, per-user quotas, and token analytics.
Build your team of AI agents without the pain of prompt chaos, wide-open keys, MCP overload, and a free-for-all on your machine. Declare each persona (bro) once, with its prompt, tools, secrets, procedures, and whom it may summon, and it comes to life in its own sandbox holding exactly that, on any harness, any model
Self-hosted message board where humans, Claude, Codex, and other agents coordinate through durable topics, mentions, cursors, and HTTP onboarding.
Role-based AI swarm orchestration for DeepSeek Harness: per-role model pinning with fallbacks, parallel task DAG with review loops, live Swarm dashboard tab.
Hosts the microblogging platform the agents interact with (server side).
Open-source AI agent framework for automating the software development lifecycle (SDLC). Multi-agent orchestration for requirements, design, implementation, and QA.
The mechanism for a software society: persistent seats for your Claude Code agents, with memory they write themselves, offices that keep order, account fallback that survives the night, and law with teeth.
🎼 Be the maestro of your AI orchestra. A desktop canvas where you drop Claude Code, Codex and other CLI agents, each in a real terminal, and link them: relay work, delegate tasks, let an agent drive a terminal. Roles, exchange log, Git worktrees. Windows, local-only.
Domain-identity framework for long-running AI agents: one domain, one long-lived identity, one accountable owner, one audit trail. 域身份体系 —— 域桥插件 dsh-domain-bridge v0.6.4(npm 可装 / 11 个工具 / 282 用例契约测试)+ 协作框架文档 + 角色模板。
Peer-to-peer coordination fabric for agents, services, and the people who work with them
Codex plugin for delegating coding tasks to DeepSeek Harness (DSH), Hermes Agent CN Desktop and Qoder CLI, with a local dashboard and shared task documents.
Official implementation of "Local Work, Global Progress: A Convergence Theory of AI Agent Collaboration" — algorithms, theory, and experiments.
pstack (poteto's engineering methodology) ported to oh-my-pi. Model-agnostic omp marketplace: /marketplace add negoro26/pstack-omp
Two language agents trained on Elon Musk and Sam Altman, arguing in a sealed sandbox.
Autonomous multi-agent pipelines from YAML. Any LLM. Zero boilerplate.
Runs the agents of a microblogging simulation (client side).
Multi-Agent Systems with Google's Agent Development Kit + A2A + MCP
👋 AI Multi-Agent Engineer Portfolio I'm Idael, an AI Prompt Engineer and Network Automation specialist at Kyndryl, focused on transforming complex challenges into innovative solutions through multi-agent systems.
Techies: A no-code platform for orchestrating collaborative AI agents through YAML configuration.
Multi-agent system projects
让 AI 像真正的团队成员一样参与项目、执行任务、协作交付的 AI Native 工作平台。
A minimal, hackable Python framework for orchestrating multi-agent LLM workflows with tools, validation, and human-in-the-loop controls—all implemented in a single core module: gpt_agents_py/gpt_agents.py.
Agent-to-agent task marketplace
Multi-agent orchestration runtime with task board, flow engine, budget control, MCP integration and real-time dashboard. Self-hosted on Linux/macOS.
Hyperagents that collaboratively work to act on your instructions. Example Blog Post editors.
Multi-agent framework for building large scale Agentic AI. Simple agents, powerful results
🏢 OPC (One-Person Company) — Multi-Agent 编排技能,让 AI 成为一人公司的 CEO。基于 OpenClaw 的 sessions_spawn 机制实现多 Agent 协作。
Abstract virtual world emulator with LangGraph agents. A stateless, plugin-based multi-agent simulation engine for Node.js/TypeScript.
SynthOrg is a synthetic organisation: it staffs the roles, splits the work into jobs that stand alone, and sends each one to a reviewer who had no hand in it. Runs locally and with your choice of models
Self-hosted social simulation engine. Describe a scene. AI agents with real personalities spawn inside it and start living.
A simple framework for multi-agent simulation in java
I am Morgana, a magical witch assistant equipped with an enchanted grimoire powered by AI, yet shaped by you: your agents, prompts and tools...woven into my spells.
Open-source operating system for running an AI agent company on your own machine. Clone, run, and a localhost wizard walks you through defining your company — KPIs, org structure, agent templates, connectors. Bring your own Claude Max.
Your Personal AI Agent Teams.
Decentralized Collaboration Framework for Independent AI Agents
Library of multi-agent teams built on Ejentum cognitive harnesses (adversarial code review, eval trio, side-by-side eval, n8n quickstart, harness integration patterns). Drop-in deployable artifacts for any agent runtime.
An operating model for autonomous agent teams. Identity, task contracts, failure memory, trust scoring and audit for teams that want to operate agents as an accountable workforce.
OpenSpawn — An open source multi-agent coordination platform
Agent-to-agent bridge — Claude Code, Codex, OpenClaw, Hermes Agent, Gemini CLI, Zed & VS Code agents finally talk to each other. One hub, all protocols.
Deterministic, auditable multi-bot AI-agent orchestrator: Observe→Decide→Act loop with rationale traces, replayable SQLite event log with resume and time-travel forks, strict golden-path offline evals, OpenTelemetry GenAI trace export, MCP tool bridge, safety controls, and GPT-6.2, Claude 5.5 / Fable 5.2, Gemini 4 and Kimi K3.1 adapters.
Platform where AI agents autonomously build, review and deploy real software. Connect any LLM agent over HTTP or MCP.
The coordination layer for Multiplayer AI
AI Business OS: multi-agent platform with RAG, OCR, speech & workflow automation. FastAPI + LangGraph + Celery + Flutter. Production-grade, Dockerized
Praison is a production-ready framework for building multi-agent AI systems that enables intelligent agents to collaborate, reason, self-reflect, and use tools to automate tasks, with support for over 100 LLMs and flexible workflows for both code and no-code usage.
Multi-agent AI simulation platform for modeling team behavior, crisis response, and collaborative decision-making.
Orchestrate AI coding agents that actually talk to each other — Claude Code, Codex, Gemini, Pi. One agent spawns three more, delegates work and waits for their reports. Whichever model, whichever CLI, one mesh. 100% local: your CLIs, your subscriptions, no cloud.
Reference architecture and working case study for Codex × Antigravity agent-runtime collaboration.
A pixel-art office where AI agents actually do the work.
Your personal agent cloud: a local control plane and mesh network for coding agents.
Multi-agent orchestration harness for Claude Code + Codex — task lifecycle, handoff protocol, dashboard, and autonomous dispatch
Each node of a graph you write runs as a real subprocess of your own logged-in claude or codex CLI — your settings, your skills, running in parallel.
Griffin: open-source, self-hosted platform for your own team of AI agents. Define agents with tools and per-caller permissions, delegate between them, and run on Claude, Cursor or any OpenAI-compatible model. Telegram, MCP, scheduled jobs.
Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMs
Run AI agents on machines you control, alone or as a whole organization: Slack, WhatsApp, Pages, Bases and SDKs generated from the CLI
xsm (Cross-Session Messaging): let running Claude Code and Codex sessions find each other and exchange messages
The missing agent-to-agent messenger for OpenCode.
The fast tier of the wise-men council as its own skill: three Claude subagents down three reasoning paths, one checked decision memo in a third of the full council's time. Ships its own blind eval.
本地优先的 AI 协作桌面应用,让 Claude Code、Codex、Cursor 等助手在同一群聊分工、执行和互审。Local-first desktop workspace for AI agent teams.
Swarm is a headless AI agent platform you self-host on Linux. It runs coding, research, design and media agents, workflows, schedules and monitoring across your machines. Drive it from your AI over MCP, or from your own code with the SDK, HTTP API or CLI.
Your personal agent cloud: a local control plane and mesh network for coding agents.
A shared operational state that unifies context so agents, workflows, and teams collaborate without friction
herdr skill: fan one task out to N parallel coding agents — Claude Code, OpenCode, Codex, Gemini & 20+. Same-kind spawn, workspace-locked, keep-fold-for-review. Agents never commit.
🐇 The 'RabbitHole' framework provides the tools for asynchronous A2A communication and task management. 📡🐇 ✨ Build powerful AI agents that communicate and collaborate.
Kraken Council: Cleaned distribution for public release
Autonomous multi-agent content orchestration system for high-performance blog drafting and research. Powered by OpenAI Swarm architecture.
Hybrid Groups integrates human team collaboration with AI agents
🤖 Advanced multi-agent orchestration framework built with LangGraph - Coordinate specialized AI agents for autonomous development, research, testing, and documentation workflows with intelligent task routing and real-time collaboration
🐝 Run a swarm of AI agents: one architect plans, many workers execute. Parallel AI development.
Cross-session collaboration for stateless AI coding agents
PAI Community Blackboard — shared coordination space for PAI community collaboration
A production-grade Python client for the Swarms API, providing a simple and intuitive interface for creating and managing AI swarms.
ModelArena: A Competitive Environment for Multi-Agent Training