/ChatGPT-wechat-bot

ChatGPT for wechat

Primary LanguageTypeScriptMIT LicenseMIT

ChatGPT-wechat-bot🤖

Version License: ISC

Get a WeChat robot 🤖 based on ChatGPT in a few step. 中文文档 | English

Support

  • Support conversation.
  • Can set keywords to reset the previous conversation.
  • Support replying message when mentioning your bot in the group.
  • Set the keywords to wake up the WeChat robot in the group.
  • Support Dockerfile to deploy.
  • Support set retry times when request ChatGPT.
  • Catch conversation error and retry.
  • Other

Default config

{
  // Fill up with session token
  chatGPTSessionToken: '',
  // Setting the number of retries when API error occurs
  retryTimes: 3,
  // Setting keyword to wake up in group chat
  groupKey: '',
  // Setting keyword to wake up in private chat
  privateKey: '',
  // Setting keyword to reset context
  resetKey: 'reset',
  // Enabling auto replies from ChatGPT
  autoReply: true,
  // Using regular expression to automatically pass friends verification
  friendShipRule: /chatgpt|chat/,
}

Use with docker (support ARM)

// build
docker build --pull --rm -f "Dockerfile" -t wechatbot:latest "."

// Linux ARM
docker build --pull --rm -f "Dockerfile" -t wechatbot:latest "." --platform linux/amd64

// run, and then you will see some logs
docker run --name wechatbot wechatbot:latest

How to start?

  1. Firstly, you should have an OpenAI account, then follow the steps below to get your token.

    Refer to this article: https://github.com/transitive-bullshit/chatgpt-api.

To get a session token:

  • Go to https://chat.openai.com/chat and log in or sign up.
  • Open dev tools.
  • Open Application > Cookies. image.png
  • Copy the value for __Secure-next-auth.session-token and save it to your environment.

  1. Now, you should fill your Token value into the ChatGPTSessionToken in the directory src/config.js， then run this project on local.

    If necessary, configure other customization variables in src/config.js.

// install dependencies
npm i

// start:
// dev
npm run dev

//or
// build
npm run build
// run lib
node lib/bundle.esm.js

  1. you can see your logs, and scan qrcode to log in wechat.

    image.png

  2. Send a message to your WeChat and you will receive a reply from ChatGPT.

    image.pngimage.png

QA

  1. If your WeChat cannot log in Please check the root directory of your project, whether there is a file —— WechatEveryDay.memory-card, if so, please delete it and try it again.

  2. This package is ESM-only. It supports: Node.js >= 16.8

  3. how to get all response? you can say "continue".

image

👏🏻Contributions

Welcome to contribute your code and ideas🍵.