Tgpeetees.js

Tgpeetees.js is a framework based on Telegraf.js designed for creating GPTs in Telegram. It leverages the capabilities of the OpenAI GPT models to enable advanced conversational bots within the Telegram platform.

Table of Contents

• Installation
• Usage
• API Documentation
  • Constructor
  • init
  • addHelp
  • addBotAction
  • addBotCommand
  • botTyping
  • getUserId
  • sendToGpt
  • addChatGpt

Installation

To install Tgpeetees.js, use npm:

npm install telegraf openai tgpeetees

Usage

Here is a basic example of how to use Tgpeetees.js:

import { Tgpeetees } from 'tgpeetees';

const bot = new Tgpeetees({
  botToken: 'YOUR_TELEGRAM_BOT_TOKEN',
  openaiApiKey: 'YOUR_OPENAI_API_KEY',
  model: 'gpt-4'
});

bot.addHelp('This is a help message');

bot.addBotCommand({
  name: 'start',
  callback: (ctx) => ctx.reply('Hello, I am your GPT bot!')
});

bot.init();

API Documentation

Constructor

constructor(params: {
  botToken: string,
  openaiApiKey?: string,
  model?: string,
})

• botToken: string - Your Telegram bot token.
• openaiApiKey: string (optional) - Your OpenAI API key.
• model: string (optional) - The OpenAI model to use (default is gpt-4).

init

public async init()

Initializes and starts the bot. It also sets up graceful stop handlers.

addHelp

public async addHelp(msg: string, keyboard?: any)

Adds a help message that is shown when the /help command is used.

• msg: string - The help message to display.
• keyboard: any (optional) - Custom keyboard layout.

addBotAction

public addBotAction(action: {
  name: string,
  callback: any,
})

Adds a custom action for the bot.

• name: string - The name of the action.
• callback: any - The callback function to execute when the action is triggered.

addBotCommand

public addBotCommand(command: {
  name: string,
  callback: any,
})

Adds a custom command for the bot.

• name: string - The name of the command.
• callback: any - The callback function to execute when the command is used.

botTyping

public async botTyping(ctx: any)

Sends a "typing" action to the chat to indicate the bot is processing.

• ctx: any - The context object from Telegraf.

getUserId

public getUserId(ctx: any)

Gets the user ID from the context.

• ctx: any - The context object from Telegraf.

sendToGpt

public async sendToGpt(messages: any[], params?: any)

Sends messages to the GPT model and returns the response.

• messages: any[] - The array of messages to send to the GPT model.
• params: any (optional) - Additional parameters for the GPT model.

addChatGpt

private async addChatGpt(token: string)

Sets up the OpenAI integration.

• token: string - Your OpenAI API key.

Example

Here is a more detailed example showing how to create a GPT-enabled Telegram bot:

import { Tgpeetees } from 'tgpeetees';

const bot = new Tgpeetees({
  botToken: 'YOUR_TELEGRAM_BOT_TOKEN',
  openaiApiKey: 'YOUR_OPENAI_API_KEY',
  model: 'gpt-4'
});

bot.addHelp('This bot can chat with you using OpenAI GPT-4.');

bot.addBotCommand({
  name: 'start',
  callback: async (ctx) => {
    await bot.botTyping(ctx);
    ctx.reply('Hello, I am your GPT bot!');
  }
});


bot.init();