Verified project record

agentscope-ai/QwenPaw

qwenpaw is a personal AI assistant that runs locally or on the cloud with extensible skills, multi-agent capabilities, and multi-channel integration. It preserves data privacy by keeping data on the user's machine without third-party hosting or data upload.

Project overview

It bundles its own local runtime models and requires no paid services while supporting kernel-level execution isolation across major desktop operating systems, making private AI assistance practical without API dependencies.

Project type: AI Agent · AI Coding
Use cases: Chat Assistants · Coding & Development
Deployment: Refer to project documentation
License: Apache-2.0

Best for

General users and creators who want a personal AI assistant that runs locally without coding or paid cloud services.
Users who need data to stay on their own machine with no third-party hosting or upload.

Key capabilities

Built-in local runtime using QwenPaw-Flash models or integrations with Ollama and LM Studio to run LLMs without an API key.
Execution isolation using Seatbelt (macOS), Bubblewrap or Landlock (Linux), and AppContainer (Windows).
Support for a Chrome extension for browser automation.

Limitations and risks

Although a sandbox and guards exist, sophisticated command injection or reverse shell attacks could pose risks.
The macOS app is not notarized and requires users to bypass Gatekeeper security restrictions.
Script installation may not work in restricted network environments or corporate firewalls.

Getting started

Install via the script installer or pip, initialize with qwenpaw init --defaults, then start the app with qwenpaw app. Cross-platform desktop apps are also available without Python configuration.

Evidence and sources

GitHub project description: Your Personal AI Assistant; easy to install, deploy on your own machine or on the cloud; supports multiple chat apps with easily extensible capabilities.
README: Three-layer memory — live working context, full verbatim history, and distilled knowledge. Older turns evict but stay recallable on demand; nothing is summarized away or lost.
README: Deploy locally — data stays on your machine. No third-party hosting, no data upload.
README: Then open the Console in your browser at **http://127.0.0.1:8088/** to configure your model.
README: No Python setup required, one command installs everything.

Reviewed at: 2026-07-18T18:39:40Z