WhatsApp Bot with Groq API

This project is a simple WhatsApp bot that uses the Groq API to generate AI-powered replies. It is built with Flask and can be deployed on any server that supports Python 3.8+.

Features

• Receives WhatsApp messages via Meta's webhook
• Replies using Groq's LLaMA models (model: llama3-8b-8192 by default)
• Handles both incoming messages and status updates (delivered/read)
• Easy .env-based configuration

Requirements

• Python 3.8 or newer
• WhatsApp Cloud API access (Meta for Developers)
• Groq API key (get one here)

Setup

1. Install dependencies:

python -m pip install -r requirements.txt

2. Create a .env file in the project root with the following keys:

WHATSAPP_TOKEN=your_whatsapp_access_token
PHONE_NUMBER_ID=your_phone_number_id
VERIFY_TOKEN=your_webhook_verify_token
GROQ_API_KEY=your_groq_api_key

• WHATSAPP_TOKEN: Get from your Meta for Developers dashboard (WhatsApp Cloud API > API Token).
• PHONE_NUMBER_ID: Also from Meta dashboard (WhatsApp > Phone Numbers).
• VERIFY_TOKEN: Any random string you choose; must match the one set in your Meta webhook config. You can generate a secure token in Python:

  import secrets
  print(secrets.token_hex(16))

  Copy the output and set it as your VERIFY_TOKEN in .env.
• GROQ_API_KEY: Get from Groq Console (requires a Groq account).

Development Workflow

• Use worker.py to automatically restart the bot when you edit main.py during development. This script watches for changes and reloads the bot, so you don't have to restart manually.

  python worker.py

  Press Ctrl+C to stop the watcher and the bot.

How the Bot Uses Groq

• The bot uses the llama3-8b-8192 model by default for fast, cost-effective, and concise WhatsApp replies. You can change the model in main.py (see the generate_ai_reply function) to other Groq models like llama3-70b-8192 for more advanced responses (may be slower or costlier).
• The system prompt is set in code to instruct the LLaMA model to be concise, helpful, and friendly, and to keep replies under 2-3 sentences. This prompt is always sent before the user's message, so the bot's tone and style are consistent for every chat.
• You can customize the prompt or model in main.py to fit your use case.

3. Run the bot:

   python main.py

4. Expose your server to the internet (for Meta webhook):

   • Install ngrok if you don't have it:
     • Download from https://ngrok.com/download
     • Or install via package manager (see ngrok docs)
   • Start a tunnel to your Flask server:

     ngrok http 5000

   • Copy the HTTPS forwarding URL from ngrok's output (e.g., https://abc123.ngrok.io).
   • Go to your Meta for Developers App Dashboard, select your app, and navigate to WhatsApp > API Setup > Webhooks.
   • Set the webhook URL to https://<your-ngrok-subdomain>.ngrok.io/webhook and use your VERIFY_TOKEN from .env.

Libraries Used

• Flask - Web server
• python-dotenv - Loads environment variables
• requests - HTTP requests
• groq - Groq API client

Where to Get API Keys & Setup Guides

• WhatsApp Cloud API:
  • Follow the official Meta guide: Get Started with WhatsApp Cloud API
  • This covers creating a Meta developer account, setting up an app, getting your WhatsApp token, phone number ID, and configuring webhooks.
• Groq API:
  • Go to Groq Console
  • Sign in, create an API key, and copy it to your .env as GROQ_API_KEY.

Notes & Troubleshooting

• This bot only handles text messages. You can extend it to handle media or interactive messages by modifying the webhook handler in main.py.
• For production, always use HTTPS and keep your tokens/keys secret.
• In the Meta developer dashboard, use the provided test number for development. Using your real number may result in a ban.
• If you see errors about missing keys or tokens, double-check your .env file and restart the bot.
• If you get a Groq configure error, your API key may be invalid or expired.
• If the bot only echoes messages, your Groq API key is missing or not loaded.

How to generate longer lasting tokens

• Go to Meta Business Dashboard

• Open Meta Business Settings .

Make sure you’re inside the correct Business Account where your WhatsApp number lives.

Check User Roles

Navigate to Accounts → WhatsApp Accounts.

Select the correct WhatsApp account.

Ensure your system user (or yourself) has Admin access to this account.

⚠️ Without proper roles, you won’t be able to generate a valid token.

Create a System User (if not already)

Go to Users → System Users.

Add a new system user (give it a name like whatsapp-bot-user) and assign Admin role.

Generate a token for this system user.

While generating, choose the App linked to your WhatsApp account.

Assign Assets & Permissions

In System User → Assign Assets, attach your WhatsApp account.

Grant the following permissions:

whatsapp_business_messaging

whatsapp_business_management

Save changes.

Generate a Token

In the System User section, click Generate Token.

Pick the right App.

Select required permissions (whatsapp_business_messaging, whatsapp_business_management).

Choose Never Expire if available (some system user tokens allow this).

Copy the token.

Use the Token in Your App