Mcpwhatsapp

🚀 WhatsApp MCP Assistant (Client-Server)

This project constructs a client-server pipeline to send and schedule WhatsApp messages via natural language. The server uses Mistral's large language model to interpret user intent and invokes Twilio to actually execute WhatsApp messaging.

🚀 Quick Start

Files

• mcp_server.py: The main MCP server file that uses Mistral for intent parsing and calls Twilio to send messages.
• mcp_client.py: A simple CLI client for interacting with the server through natural language.
• config.py: Stores API keys and constants, including phone number configurations.

Server-Client Concept

The server uses Mistral's large language model to parse the user's intent and convert it into executable instructions (such as sending a message). The client is a CLI interface where users can enter prompts like "Send a message to Nir", and the backend automatically handles the subsequent operations.

Twilio

Twilio provides a programmable WhatsApp messaging API. In this project, it is used to actually send messages or simulate scheduling on WhatsApp.

⚠️ WhatsApp 24-Hour Window Limit (Twilio)

If you are using Twilio's WhatsApp API, please note that free-form messages (i.e., ordinary text messages) are only allowed to be sent within 24 hours after the user last communicated with your WhatsApp number.

If you try to send a message outside this window, you will see this error on the Twilio dashboard: "Cannot send free-form messages because you are outside the allowed time window. If you are using WhatsApp, use a message template."

✅ Solution

• You must use a pre-approved message template to send messages outside the 24-hour window.
• You can create message templates in Twilio Console > Messaging > Templates.

💡 When Testing Locally

If you are just testing on your own number and encounter problems, make sure you have recently sent a message from your WhatsApp to the Twilio number - this will reset the 24-hour window.

MCP Concept

The MCP server encapsulates a large language model (Mistral) with function call capabilities to dynamically route user intent to appropriate tools (sending/scheduling WhatsApp messages). The client allows prompt-based CLI input.

📦 Installation

1. Install dependencies:

   pip install twilio requests fastapi
   

2. Update config.py with real values.
3. Run the server:

   python mcp_server.py
   

4. Run the client in another terminal:

   python mcp_client.py