Build a Telegram Bot with Python Telebot: A Step‑by‑Step Guide

0 Data Architecture

+-----+
        | TTS |
        +--+--+
           |
           v
     +------------+
     | Server api |
     +-----+------+
           |
     +-----v-----+
     |   Agent   |
     +-----+-----+
       |       |
   +---v---+ +-v------+
   | tools | | memory |
   +-------+ +--------+

Data flows from memory upward to TTS and then downward to tools.

1 Obtain Bot Token

Search for BotFather on Telegram and choose the officially verified (blue‑check) account; many impostors exist.

This is the genuine one.

1.1 Create a New Bot

1.2 Edit Bot Information

Edit the "About" field:

Set Bot Avatar

Now you can directly access the bot.

2 Install the telebot Package

import telebot
# previously obtained user token
bot = telebot.TeleBot('xxx:xxx')

3 Write Client Code

@bot.message_handler(commands=['start'])
def start_message(message):
    bot.send_message(message.chat.id, '你好我是JavaEdge，欢迎光临!')

python tele-qwen.py

Run the script; the bot will start responding.

3.1 Simple Reply

3.2 Reply with Reference

@bot.message_handler(commands=['start'])
def start_message(message):
    bot.reply_to(message, '你好!')

4 Connect Bot to a Server‑Side LLM API

Link the bot to a chat interface that forwards user queries to a local LLM service.

@bot.message_handler(func=lambda message: True)
def echo_all(message):
    # bot.reply_to(message, message.text)
    try:
        encoded_text = urllib.parse.quote(message.text)
        response = requests.post('http://localhost:8090/chat?query=' + encoded_text, timeout=100)
        if response.status_code == 200:
            ai_say = json.loads(response.text)
            if "msg" in ai_say:
                bot.reply_to(message, ai_say["msg"]["output"])
                audio_path = f"{ai_say['id']}.mp3"
                asyncio.run(check_audio(message, audio_path))
            else:
                bot.reply_to(message, "对不起,我不知道怎么回答你")
    except requests.RequestException as e:
        bot.reply_to(message, "对不起,我不知道怎么回答你")

async def check_audio(message, audio_path):
    while True:
        if os.path.exists(audio_path):
            with open(audio_path, 'rb') as f:
                bot.send_audio(message.chat.id, f)
            os.remove(audio_path)
            break
        else:
            print("waiting")
            await asyncio.sleep(1)

bot.infinity_polling()

This forwards the LLM's textual and audio responses back to Telegram users.

Reference: Telegram Bot API.