QwenPaw

QwenPaw

3.0

Personal AI agent workstation — self-hosted, connected to 10+ messaging channels, extensible with Skills and MCP.

Last Updated Nov 10, 2026
QwenPaw screenshot

About

QwenPaw is a personal AI agent workstation that runs in your own environment — local or cloud. It connects your agent to more than 10 messaging channels, including DingTalk, Feishu, QQ, Discord, and Telegram, so your agent lives where you and your team already talk. The system is extensible with Skills and MCP, supporting multi-agent collaboration and memory management — a framework for building an agent that is genuinely yours rather than a tenant of someone else's SaaS.

Key Features

**Self-hosted deployment: ** run locally or in your own cloud — your agent, your environment, your data.
**10+ messaging channels: ** DingTalk, Feishu, QQ, Discord, Telegram, and more as native interfaces.
**Skills extensibility: ** add capabilities to your agent as composable skills.
**MCP support: ** connect tools and context through the Model Context Protocol.
**Multi-agent collaboration: ** coordinate multiple agents, with memory management to keep context durable.

Pricing & Fees

  • Pricing: Not publicly listed on the pages we reviewed; check the official site for current plans.
  • Deployment: Runs in the user's own environment, with support for both local and cloud deployment.
  • Channel support: Connects to 10+ channels including DingTalk, Feishu, QQ, Discord, and Telegram.
  • Extensibility: Supports Skills and MCP for multi-agent collaboration and memory management.

Who It's For

QwenPaw is for developers and technical teams building a personal or work assistant they want to own end to end, organizations standardized on DingTalk or Feishu that want agents in their existing chat, and tinkerers who want a multi-agent, memory-backed setup under their own roof.

FAQ

Where does it run?In your own environment — local or cloud deployment.

Which chat platforms does it connect to?10+ channels including DingTalk, Feishu, QQ, Discord, and Telegram.

How is it extended?Through Skills and MCP, with multi-agent collaboration and memory management built in.