CoPaw - AI Pet Assistant

Can I run CoPaw entirely on my local machine without sending data to external servers?

Yes, CoPaw is designed for local execution. You can run local LLMs via llama.cpp or MLX (for Apple Silicon), handle all message processing locally, and keep your data on your own hardware. You can also deploy it to your own cloud infrastructure if you prefer, giving you complete control over where your data lives.

Which chat platforms can CoPaw connect to?

CoPaw integrates with DingTalk, Feishu, QQ, Discord, and iMessage. It uses a message queue for reliable multi-channel processing, so you can receive and respond to messages across these platforms from a single local instance.

How do I add new skills or automate custom workflows in CoPaw?

CoPaw has a modular skill system with a CLI for management. You can write new skills using its extensible architecture, which decouples Prompt, Hooks, Tools, and Memory components. Skills support cron scheduling, and CoPaw's heartbeat mechanism lets you trigger autonomous tasks on a schedule.

What's the easiest way to get CoPaw running?

You can install CoPaw via pip, Docker, or curl, or use a one-click cloud deployment option. The low barrier to entry means you can have it up and running quickly without complex infrastructure setup, whether you're deploying locally or in the cloud.