CoPaw (Alibaba)

Alibaba Cloud's open-source AI agent supporting DingTalk, Feishu, and more. End-cloud dual-mode, 2.4k+ GitHub stars. Free and open source.

Alibaba's Open-Source AI Agent Platform

CoPaw (Co Personal Agent Workstation) is Alibaba Cloud Tongyi Laboratory's open-source AI agent framework developed by the AgentScope team. Launched in early 2026, it's designed as a comprehensive Personal Agent Workstation for enterprise and individual users with support for multiple Chinese and international IM platforms.

Core Philosophy: "A warm little paw that is always ready to help"

Launch Date: Early 2026
Company: Alibaba (AgentScope team)
GitHub Stars: 2,400+
License: Apache-2.0

In March 2026, Alibaba also launched Wukong (悟空), an AI-native enterprise platform that brings advanced agentic AI capabilities directly into business workflows. Wukong represents Alibaba's enterprise-focused AI agent solution, while CoPaw serves as the open-source personal assistant framework.

Wukong Features:

Coordinates multiple AI agents for complex business tasks
Document editing, spreadsheet updates, meeting transcription, research
Integration with 20+ million DingTalk users
Operates across local systems, web browsers, and cloud platforms
Currently in invitation-only beta testing
Plans for Slack, Teams, and WeChat integration

Why CoPaw?

🤖 Open-Source Personal AI Workstation

CoPaw is not just another chatbot wrapper—it's a comprehensive Personal Agent Workstation designed to:

Live where you work
Remember what you say
Operate entirely under your control
Evolve and adapt over time

"Soulful" Design: Possesses a defined personality and memory that evolves with use.

🌐 Multi-Platform Support

Native support for multiple messaging platforms:

DingTalk (Alibaba's enterprise platform)
Feishu (ByteDance's collaboration platform)
Discord (International community platform)
iMessage (Apple's messaging)
QQ (Tencent's messaging platform)
And more

🏠 End-Cloud Dual-Mode

Choose your deployment model:

Local Deployment: Run on your own machine for complete privacy
Cloud Deployment: Deploy to cloud for 24/7 availability and scalability

🧠 Reme Memory Engine

Advanced memory system for maintaining context across:

Multiple conversations
Different sessions
Long-term interactions
Cross-platform continuity

🔓 Open Source & Free

Apache-2.0 License: Completely free to use and modify
No Subscription: No recurring fees or API middlemen
Community-Driven: Active development by Alibaba ecosystem
Full Control: You own the instance and the data

Who Should Choose CoPaw?

Your Profile	Recommendation
Open-Source Advocate	✅ Apache-2.0 license, fully auditable and self-hostable
Multi-IM Platform User	✅ Broadest IM support: DingTalk, Feishu, Discord, iMessage, QQ
Enterprise Developer	✅ Alibaba-backed, enterprise-grade architecture
Non-Technical User	❌ Requires Python setup, steeper learning curve
WeChat-First User	❌ No native WeChat integration. Consider QClaw
Quick-Start Seeker	❌ More complex than one-click alternatives

If you value open-source freedom and need broad IM platform support, CoPaw is the strongest open-source contender. For native WeChat integration, consider QClaw.

Key Features

$#1.
Multi-Platform Integration

DingTalk (Alibaba's Slack competitor)
Feishu (ByteDance's collaboration platform)
QQ (Tencent's messaging platform)
Discord and more

$#2.
Reme Memory Engine Advanced memory system for maintaining context across conversations and sessions.

$#3.
Tongyi Qianwen Model Powered by Alibaba's Tongyi Qianwen large language model, optimized for Chinese and English.

$#4.
End-Cloud Architecture Choose between local deployment (privacy) or cloud deployment (scalability).

Installation

# Clone repository
git clone https://github.com/agentscope-ai/CoPaw
cd CoPaw

# Install dependencies
pip install -r requirements.txt

# Run locally
python main.py

# Docker deployment
docker build -t copaw .
docker run -d --name copaw -p 8080:8080 copaw

# Cloud deployment on Alibaba Cloud
# See documentation for cloud deployment options

System Requirements

Component	Minimum	Recommended
CPU	2 cores	4 cores
Memory	1GB RAM	4GB RAM
Storage	10GB	50GB
Python	3.8+	3.10+
Docker	Optional	Recommended

Use Cases

DingTalk Automation

Automate tasks within Alibaba's DingTalk enterprise collaboration platform.

Multi-Platform Management

Manage AI assistants across multiple IM platforms from a single system.

Enterprise Knowledge Management

Build enterprise knowledge bases with intelligent retrieval and generation.

Comparison

Feature	CoPaw	AutoClaw	QClaw	OpenClaw
DingTalk	Native	No	No	Plugin
Feishu	Native	Native	No	Plugin
Open Source	Yes	No	No	Yes
Python	Yes	No	No	No
Price	Free	29-249 yuan	TBD	Free + API

Community and Support

GitHub: https://github.com/agentscope-ai/CoPaw
Website: https://copaw.bot
Documentation: https://copaw.agentscope.io/docs
GitHub Stars: 2,400+

For enterprise users, Alibaba also offers Wukong, launched March 18, 2026:

AI-native enterprise platform
Multi-agent coordination for complex workflows
20+ million DingTalk user integration
Slack, Teams, WeChat integration planned
Invitation-only beta

Sources

License

Apache-2.0 - Free and open source for commercial and personal use.

Summary

CoPaw is Alibaba's open-source AI agent framework, launched early 2026:

Open Source & Free -- Apache-2.0 license, no subscription fees
Multi-Platform -- DingTalk, Feishu, Discord, iMessage, QQ support
End-Cloud Dual-Mode -- Local or cloud deployment options
Reme Memory Engine -- Advanced context and memory management
Tongyi Qianwen Model -- Alibaba's powerful LLM
Personal Workstation -- Comprehensive AI assistant framework
Alibaba Backing -- Active development by AgentScope team

Suitable For:

✅ Users wanting open-source solutions
✅ DingTalk/Feishu enterprise users
✅ Privacy-conscious users (local deployment)
✅ Developers and technical users
✅ Multi-platform IM management
✅ Users wanting full control over their AI

Quick Start Guide

Get CoPaw (Alibaba) up and running in about 10 minutes.

Option A: Docker (Recommended)

docker pull copaw/copaw:latest
docker run -d --name copaw copaw/copaw:latest

Option B: Manual Installation

git clone https://github.com/agentscope-ai/CoPaw.git
cd copaw
npm install
npx copaw setup

Step 3: Configure Your AI Model

Set your preferred AI model and API key in the configuration file.

Step 4: Connect a Platform

Link your messaging platform (Telegram, Discord, etc.) and start sending commands.

Time to first result: ~10 minutes with Docker, ~20 minutes with manual setup.

Full documentation: https://copaw.agentscope.io/docs

Source code: https://github.com/agentscope-ai/CoPaw

FAQ

Is CoPaw (Alibaba) free to use?

Yes, CoPaw (Alibaba) is free and open source (Apache-2.0 license). You only pay for AI model API costs if using external models.

What are the system requirements for CoPaw (Alibaba)?

CoPaw (Alibaba) requires 1GB RAM of RAM minimum. Runtime: Python 3.8+, optional Docker. It runs on Windows, macOS, and Linux.

Can I self-host CoPaw (Alibaba)?

Yes. CoPaw (Alibaba) is open source (Apache-2.0) and can be self-hosted on your own hardware. Clone the repository from GitHub and follow the installation guide.

What platforms does CoPaw (Alibaba) support?

CoPaw (Alibaba) supports: Discord, Dingtalk, Feishu, Qq. Check the official documentation for the most up-to-date list of supported platforms.

How does CoPaw (Alibaba) compare to OpenClaw?

CoPaw (Alibaba) offers a different approach compared to OpenClaw. While OpenClaw provides the largest ecosystem with 13,729+ skills and maximum flexibility, CoPaw (Alibaba) focuses on chinese tech giants. Choose CoPaw (Alibaba) if you prioritize its specific features; choose OpenClaw for the broadest compatibility and community support.

Is CoPaw (Alibaba) suitable for beginners?

CoPaw (Alibaba) requires some technical knowledge to set up (Python 3.8+, optional Docker). If you are a beginner, consider starting with QClaw (one-click install) or MaxClaw (cloud-based, no setup) first, then graduate to CoPaw (Alibaba) as you gain experience.

OpenClaw - The original open-source AI agent framework with 364k+ stars
QClaw - Tencent's one-click AI agent with Hermes Agent kernel
AutoClaw - Zhipu AI's one-click desktop agent with GLM-5.1
Kimi Claw - Moonshot AI's cloud-native agent with K2.6 model
MaxClaw - MiniMax's cloud-hosted agent with M2.7 model

Not Recommended For:

❌ Non-technical users (requires setup)
❌ Users wanting zero-configuration solutions
❌ Users preferring commercial support

Enterprise Alternative: For enterprise users, consider Alibaba's Wukong platform (launched March 2026) with multi-agent coordination and DingTalk integration.

On this page