Skip to content

A wechat robot based on ChatGPT with no risk, very stable! 🚀

License

Notifications You must be signed in to change notification settings

fiyen/wechat-gptbot

 
 

Repository files navigation

Welcome to wechat-gptbot 👋

Version License: MIT Python Version litellm

A wechat robot based on ChatGPT with no risk, very stable! 🚀
English | 中文文档

🎤 Introduction

When I use bots based on itchat and wechaty, I often encounter the risk of account restrictions when scanning codes to log in. Refer to #158. Is there a safe way to use wechat bots? Here it is~

🌟 Features

  • Extremely Stable: Implement based on windows hook, no worry about risk of wechat account restriction
  • Basic Conversation: Smart reply for private chat and group chat, support multiple rounds of session context memory, support GPT-3, GPT-3.5, GPT-4, Claude-2, Claude Instant-1, Command Nightly, Palm models and other models in litellm
  • Image Generation: Support image generation, Dell-E only model for now
  • Flexible Configuration: Support prompt settings, proxy, command settings and etc.
  • Plugin System: Support personalized plugin extensions, you can easily integrate the functions you want

📝 Changelog

2023.07.13: Introduce plugin system to make gptbot have more possibilities and easy to expand #46. Here's the first interesting plugin: tiktok, try it and have fun! Also refer to docs to learn the usage and how to contribute~

🚀 Getting Start

Environment

Support Windows system(probably support Linux in the future based on sandbox) and Python needs to be installed at the same time

It is recommended that the Python version be between 3.8.X~3.10.X, version 3.10 is perfect

1. Clone repo

git clone https://github.com/iuiaoin/wechat-gptbot && cd wechat-gptbot

2. Install dependencies

pip install -r requirements.txt

Configuration

config.template.json in the root directory contains the configs template, you need to copy the template to create the final effective config.json

  cp config.template.json config.json

Then fill in the configuration in config.json, the following is the description of the default configuration, which can be customized according to the needs:

{
  "openai_api_key": "YOUR API SECRET KEY",             # Fill in your OpenAI API Key
  "model": "gpt-3.5-turbo",                            # ID of the model to use, support gpt-3.5-turbo, gpt-4, gpt-4-32k etc.
  "use_azure_chatgpt": false,                          # Whether use Azure OpenAI API
  "azure_deployment_id": "",                           # Azure model deployment name
  "role_desc": "You are a helpful assistant.",         # Role description as system prompt
  "session_expired_duration": 3600,                    # Session memory kept duration
  "max_tokens": 1000,                                  # Max tokens of characters for session memory
  "temperature": 0.9,                                  # Between 0 and 2. Higher values make the output more random, while lower values more focused
  "proxy": "127.0.0.1:3000",                           # Proxy client ip and port
  "openai_api_base": "",                               # api url used by openai service
  "create_image_size": "256x256",                      # Dall-E image size, support 256x256, 512x512, 1024x1024
  "create_image_prefix": ["draw", "paint", "imagine"], # Text prefix for image generation
  "clear_current_session_command": "#clear session",   # Clear current session
  "clear_all_sessions_command": "#clear all sessions", # Clear all sessions
  "chat_group_session_independent": false,             # Whether sessions of users are independent in chat group
  "single_chat_prefix": ["bot", "@bot"],               # Start conversation with "bot" or "@bot" in single chat to trigger the bot, leave it empty if you wanna make the bot active all the time
  "group_chat_reply_prefix": "",                       # Reply prefix in group chat
  "group_chat_reply_suffix": "",                       # Reply suffix in group chat
  "single_chat_reply_prefix": "",                      # Reply prefix in single chat
  "single_chat_reply_suffix": "",                      # Reply suffix in single chat
  "query_key_command": "#query key"                    # Querying the usage of the api key
  "recent_days": 5                                     # The usage in <recent_days> days
  "plugins": [{ "name": <plugin name>, other configs }]# Add the your favorite plugins
}

Running

1. Prepare

We need the specific wechat version and dll to make windows hook work.

  1. Download assets from the release
  2. Install WeChatSetup-3.2.1.121.exe and login
  3. Run the wechat-dll-injectorV1.0.3.exe
  4. Select 3.2.1.121-LTS.dll and click inject dll, you will see "Successfully injected: 3.2.1.121-LTS.dll"

2. Run command

python app.py

Voilà! Enjoy your exploring journey~

✨ Generous Backers

Thank you very much for your support, it will be my biggest motivation!

🤝 Contributing

Contributions, issues and feature requests are welcome!
Feel free to check issues page.

🙏 Show your support

Give a ⭐️ if you like this project!

📢 Announcements

The WeChatSetup is coming from wechat-windows-versions and wechat-dll-injector from wechat-bot, so you can use it without concern. Also thanks the two repo's owners for their contributions.

💖 Sponsor

Become a Sponsor on AFDIAN. Your name will be specifically listed under Generous Backers~

About

A wechat robot based on ChatGPT with no risk, very stable! 🚀

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%