zhayujie/cowagent repository overview

⁠CowAgent

Synced with zhayujie/chatgpt-on-wechat. CowAgent is a super AI assistant powered by LLMs.

CowAgent (chatgpt-on-wechat) is an AI super assistant powered by LLMs, capable of autonomous task planning, operating computers and external resources, creating and executing Skills, and continuously growing with long-term memory.

It supports flexible model switching, handles multimodal messages (text, voice, images, files), and can be integrated into WeChat, Web, Feishu, DingTalk, WeCom, QQ, and WeChat Official Account — running 24/7 on your server.

⁠Features

• Autonomous Task Planning — understands complex tasks and plans execution with multi-step tool invocation
• Long-term Memory — persists conversation memory with keyword and vector retrieval
• Skills System — install skills from Skill Hub / GitHub, or create custom skills through conversation
• Built-in Tools — file I/O, terminal, browser automation, scheduled tasks, web search, and more
• Multiple Models — OpenAI, Claude, Gemini, DeepSeek, MiniMax, GLM, Qwen, Kimi, Doubao, etc.
• Multi-platform — WeChat, Web, Feishu, DingTalk, WeCom Bot, QQ, WeChat MP

⁠Quick Start

⁠1. Get the Compose file

curl -O https://cdn.link-ai.tech/code/cow/docker-compose.yml

⁠2. Configure

Edit docker-compose.yml and set your model API key and channel type:

environment:
  CHANNEL_TYPE: 'weixin'
  MODEL: 'MiniMax-M2.7'
  MINIMAX_API_KEY: 'your-api-key'

⁠3. Run

docker compose up -d

The Web console is available at http://localhost:9899/chat.

⁠4. View logs

docker logs -f chatgpt-on-wechat

⁠Environment Variables

See all configuration options in the documentation⁠.

⁠Volumes

⁠Ports

⁠Links

• GitHub: https://github.com/zhayujie/chatgpt-on-wechat⁠
• Documentation: https://docs.cowagent.ai/⁠
• Website: https://cowagent.ai/⁠
• License: MIT