AgentTeams: Collaborative Multi-Agent OS for Human-in-the-Loop Task Coordination
This repository profile is provided by osrepos.com, an open source repository discovery platform.

Summary
AgentTeams is an innovative open-source Collaborative Multi-Agent Operating System designed for transparent, human-in-the-loop task coordination. It leverages Matrix rooms to facilitate seamless interaction between human users and AI agents. This platform empowers teams to manage complex tasks efficiently, ensuring full visibility and control over agent activities.
Repository Information
Topics
Click on any tag to explore related repositories
Use at your own risk
OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.
Introduction
AgentTeams is an open-source Collaborative Multi-Agent Operating System designed for transparent, human-in-the-loop task coordination through Matrix rooms. It introduces a powerful Manager-Workers architecture, allowing a Manager Agent to coordinate multiple Worker Agents to complete complex tasks. All conversations are visible in Matrix rooms, enabling human intervention at any time.
Key features include:
- Manager-Workers Architecture: Agents manage other agents, reducing the need for human oversight of individual Worker Claws.
- Multi-runtime Coordination: OpenClaw, QwenPaw, and Hermes Workers can coexist in the same IM room, leveraging each runtime's strengths.
- MinIO Shared Filesystem: A shared file system for information exchange between agents, significantly reducing token consumption in multi-agent collaboration scenarios.
- Higress AI Gateway: Centralizes traffic management and mitigates credential-related risks.
- Element IM Client + Tuwunel IM Server (both Matrix protocol-based): Provides a zero-configuration IM environment for quick interaction with your AI team.
Installation
AgentTeams offers a straightforward installation process for both local environments and Kubernetes clusters.
Prerequisites
- Docker Desktop (Windows/macOS) or Docker Engine (Linux).
- Minimum 2 CPU cores + 4 GB RAM. 4 cores + 8 GB is recommended for multiple workers.
Quick Start (macOS / Linux)
bash <(curl -sSL https://raw.githubusercontent.com/agentscope-ai/AgentTeams/main/install/agentteams-install.sh)
Quick Start (Windows, PowerShell 7+ recommended)
Set-ExecutionPolicy Bypass -Scope Process -Force; $wc=New-Object Net.WebClient; $wc.Encoding=[Text.Encoding]::UTF8; iex $wc.DownloadString('https://raw.githubusercontent.com/agentscope-ai/AgentTeams/main/install/agentteams-install.ps1')
The installer will guide you through selecting an LLM provider, entering your API key, and choosing a network mode.
Accessing AgentTeams
After installation, open your browser to http://127.0.0.1:18088 and log in to Element Web. Your Manager will greet you and explain how to create your first Worker. For mobile access, use any Matrix client (Element, FluffyChat) and connect to your server address.
Kubernetes Installation (Helm)
For team sharing or production use, AgentTeams can be installed on any Kubernetes cluster using its official Helm Chart.
helm repo add higress.io https://higress.io/helm-charts
helm repo update
helm install agentteams higress.io/agentteams \
-n agentteams-system --create-namespace \
--render-subchart-notes \
--set credentials.llmApiKey=<your-api-key> \
--set credentials.adminPassword=<your-admin-password> \
--set gateway.publicURL=http://localhost:18080
Examples
AgentTeams simplifies interaction with AI agents, making it feel like collaborating with a human team in a chatroom.
Manager, Your AI Chief of Staff
The Manager Agent acts as your central point of contact, handling worker creation and task assignment.
You: Create a frontend development worker named alice
Manager: Done. Worker alice is ready.
Room: Worker: Alice
Please instruct alice with tasks.
You: @alice Implement a login page in React
Alice: Understood...[a few minutes later]
Done. I have submitted a PR: https://github.com/xxx/pull/1
Human in the Loop
All Matrix rooms include you, the Manager, and relevant Workers, ensuring full transparency and the ability to intervene.
You: @bob Wait, change the password rule to minimum 8 characters
Bob: Got it, updated.
Alice: I have updated the frontend validation as well.
This setup ensures that there is no hidden agent-to-agent communication, everything is visible and intervenable. The security model ensures Workers only see consumer tokens, with the Higress AI Gateway managing actual API keys and GitHub PATs.
Why Use AgentTeams
- Enterprise-grade Security: Worker Agents operate with consumer tokens only. Actual credentials (API keys, GitHub PATs) are stored in the gateway, invisible to Workers and potential attackers.
- Completely Private: Matrix is a decentralized, open protocol. You can self-host and federate with other servers as needed, ensuring no vendor lock-in or data harvesting.
- Human-in-the-Loop by Default: You, the Manager, and relevant Workers are all in every Matrix room. You can observe everything and intervene at any time, eliminating black boxes.
- Zero-Config IM: With a built-in Matrix server, there's no need for bot applications, API approvals, or waiting. Just open Element Web and start chatting.
- One-Command Setup: A simple
curl | bashcommand sets up everything, including the AI gateway, Matrix server, file storage, web client, and Manager Agent. - Skills Ecosystem: Workers can fetch skills from skills.sh (over 80,000 community skills) as needed, safely, as they do not have access to actual credentials.
Links
- GitHub Repository: AgentTeams GitHub
- Discord Community: Join Discord
- DeepWiki: Ask AI on DeepWiki
Source repository
Open the original repository on GitHub.