hp0912/wechat-robot-client

What it solves

This project solves the need to add automation capabilities to a personal WeChat account. It provides a management backend to create robot instances that can perform AI chat, image generation, voice synthesis, group management, and more, while allowing self-hosting for full control.

How it works

The system separates a client and a server. The server uses the iPad protocol (source not public) and runs as a Docker image. The management backend has a frontend and backend, and the client connects to the server using that protocol. Users manage everything from a web-based admin panel. It supports multiple login methods (iPad, Windows, etc.) and integrates MCP protocol, a Skills engine, and a knowledge base for long-term memory.

Who it’s for

This is for developers or hobbyists who want to add AI and automated features to their own WeChat. Users should be comfortable with Docker and basic DevOps tasks, as deployment can be done locally or on a remote server.

Highlights

• Supports AI chat with models like ChatGPT, DeepSeek, and Qwen
• AI image generation (text-to-image, image-to-image) and text-to-video
• AI voice: text-to-speech and long-text-to-speech
• Group management: welcome messages, activity rankings, daily summaries, word clouds, and more
• Full MCP protocol support and built-in Skills engine
• Knowledge base engine for persistent memory in AI conversations
• View and interact with朋友圈 (Moments), auto-comment and like
• Friend management: add, verify, and auto-accept requests
• Scheduled tasks and group message monitoring with keyword alerts
• Sub-agent orchestration and sending quoted messages
• Supports multiple login modes including iPad, Windows, and Mac (with auto slider passing)