Indexed:261 news/91 AI tools/372 open-source projects/64 learning resources
HomeNewsOpen SourceModelsAI ToolsLearn
Home / Open Source / CoPaw

CoPaw

CoPaw is an open-source personal AI assistant that deploys on your machine or in the cloud, connects to DingTalk, Feishu, QQ, Discord, iMessage, and more, with extensible skills and local model support.

PythonApache-2.0Agent
⭐ GitHubhttps://github.com/agentscope-ai/CoPaw
29,103
Stars
+14,243
Star growth
Jul 28, 2026
Last updated
3,728
Clicks

📌 What is CoPaw?

CoPaw (Co Personal Agent Workstation) is an open-source, self-hosted personal AI assistant built by the AgentScope team. It is designed to be easy to install and deploy—either on your own machine or in the cloud—and to connect seamlessly with the chat apps you already use every day.

The name "CoPaw" is a play on words: it stands for Co Personal Agent Workstation, and evokes a "co-paw"—a warm partner always by your side. More than a cold productivity tool, CoPaw is meant to be a helpful companion for your digital life.

🚀 Key Features

1. Multi-Channel Support

CoPaw connects to all major messaging and collaboration platforms through a single agent:

  • DingTalk
  • Feishu
  • QQ
  • Discord
  • iMessage
  • And more via extensible channel integrations

2. Flexible Deployment

Method Description
pip install copaw Standard Python install, 3-command quick start
One-line installer No Python required—installer handles everything automatically
Docker Pull and run with a single command
ModelScope Studio One-click cloud setup with no local installation

3. Local Model Support

Run LLMs entirely on your own hardware—no API keys or cloud services required:

Backend Best for
llama.cpp Cross-platform (macOS / Linux / Windows)
MLX Apple Silicon Macs (M1/M2/M3/M4) 
copaw models download Qwen/Qwen3-4B-GGUF
copaw models   # select the downloaded model
copaw app      # start the server

4. Extensible Skills System

  • Built-in cron job scheduler for automated tasks
  • Custom skills stored in your workspace and auto-loaded at startup
  • No vendor lock-in—your skills, your data

5. Memory & Personalization

  • Context management and long-term memory under your control
  • Scheduled reminders delivered to any connected channel
  • Personal knowledge base for research and note-taking

6. Web Console UI

A clean browser-based console at http://127.0.0.1:8088/ for:

  • Direct chat with the assistant
  • Model and API key configuration
  • Channel management
  • Environment variable settings

🛠️ Quick Start

# Install
pip install copaw

# Initialize with defaults
copaw init --defaults

# Launch the console
copaw app

Then open http://127.0.0.1:8088/ in your browser.

💡 Use Cases

Category Examples
Social Daily digest of trending posts from Xiaohongshu, Zhihu, Reddit; Bilibili/YouTube video summaries
Productivity Newsletter digests pushed to DingTalk/Feishu/QQ; contact management from email and calendar
Creative Describe a goal overnight and get a draft ready by morning
Research Track AI/tech news; build a personal knowledge base
Desktop Organize local files; read and summarize documents; request files via chat
Automation Combine Skills + cron to build your own agentic workflows

🏗️ Architecture Overview

CoPaw
├── src/copaw/          # Core Python package
│   ├── agent/          # Agent logic and memory
│   ├── skills/         # Built-in and user skills
│   └── channels/       # Messaging platform integrations
├── console/            # TypeScript/React web console frontend
├── deploy/             # Docker and cloud deployment configs
├── scripts/            # Installer scripts (bash, PowerShell)
└── website/            # Documentation site

Tech Stack:

  • Backend: Python 3.10–3.14, built on AgentScope
  • Frontend: TypeScript + React
  • Local inference: llama.cpp, MLX
  • License: Apache 2.0

📦 Installation Options

pip (recommended)

pip install copaw

One-line (macOS/Linux)

curl -fsSL https://raw.githubusercontent.com/agentscope-ai/CoPaw/master/scripts/install.sh | bash

One-line (Windows PowerShell)

irm https://raw.githubusercontent.com/agentscope-ai/CoPaw/master/scripts/install.ps1 | iex

Docker

docker pull agentscope/copaw:latest
docker run -p 8088:8088 -v copaw-data:/app/working agentscope/copaw:latest

📖 Documentation

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

Topic Link
Introduction docs/intro
Quick Start docs/quickstart
Console docs/console
Channels docs/channels
Skills docs/skills
Local Models docs/local-models
Memory docs/memory
CLI docs/cli

🤝 Community & Links

⚖️ License

CoPaw is released under the Apache License 2.0 — free to use, modify, and distribute.