WeChat Bot is an open-source AI agent that connects large language models to the messaging platforms people actually use — Telegram, WhatsApp, Lark (Feishu) and WeChat — for automatic replies, group-chat assistance and contact management. It has earned about 11,400 stars on GitHub and is actively maintained in 2026.
The project started as a way to put ChatGPT inside WeChat and grew into a multi-platform IM agent. You connect the model backend of your choice — ChatGPT, Claude, Kimi, DeepSeek, a local Ollama server and others are supported — and the bot answers messages in private chats and groups according to the persona and rules you configure. Beyond simple auto-replies, it includes tools for community operation: analyzing group conversations, managing contacts and detecting inactive friends, which makes it popular with community managers and small businesses running support or interest groups on these platforms. It builds on the Wechaty ecosystem for the messaging side, so new platforms and puppet providers can be plugged in. As with all unofficial messaging automation, you should respect each platform’s terms and rate limits, but as an open, self-hosted way to experiment with LLM-powered chat agents on real messaging apps, the project is one of the most starred of its kind.
WeChat Bot is free and open source under the MIT license. You self-host it and pay only for any model API keys you connect; local models via Ollama make it fully free to run.
WeChat Bot is for community managers, small businesses and developers who want an LLM assistant living inside their messaging apps — answering questions, helping run groups and automating routine conversations on WeChat, Telegram, WhatsApp or Lark.