Skip to main content
Glama

Twilio WhatsApp FastMCP Server

by wubbyweb

Twilio WhatsApp FastMCP Server 💬

This project provides a simple FastMCP server that allows sending WhatsApp messages using the Twilio API.

Features

  • 📲 Sends WhatsApp messages via Twilio.
  • 🤖 Exposes a send_whatsapp tool for use with FastMCP clients (like AI models).
  • 🔒 Loads configuration securely from a .env file.
  • 🧪 Includes a basic test script (/home/rj/Code/mcp-generated/twilio_test.py) for direct Twilio API interaction.

Setup

  1. Get the Code: Clone this repository or download the source files into /home/rj/Code/mcp-generated/.
  2. **Create a Virtual Environment (Recommended){ @@ -77,11 +77,11 @@

2. Running the Test Script

-The /home/rj/Code/mcp-generated/twilio_test.py script provides a way to directly test sending a message using your Twilio credentials without the FastMCP server. +The /home/rj/Code/mcp-generated/twilio_test.py script provides a way to directly test sending a message using your Twilio credentials without the FastMCP server. 🛠️

  • Modify the script: Ensure the to= number in /home/rj/Code/mcp-generated/twilio_test.py is a WhatsApp number linked to your Twilio Sandbox (if using the Sandbox) or any valid WhatsApp number (if using a purchased Twilio number). The from_ number should typically be your Twilio Sandbox number (whatsapp:+14155238886) or your purchased Twilio WhatsApp number.
  • Run the script:
python /home/rj/Code/mcp-generated/twilio_test.py ``` This will attempt to send a hardcoded message ("Is this working?") from the specified Twilio number to the specified recipient.

:** bash 🌱 cd /home/rj/Code/mcp-generated/ python -m venv venv source venv/bin/activate # On Windows use `venv\Scripts\activate`

  1. Install Dependencies: The necessary dependencies are listed in /home/rj/Code/mcp-generated/whatsapp_server.py. You can install them using pip:
    pip install twilio python-dotenv pydantic-settings fastmcp
    Alternatively, if using the FastMCP framework features:
    fastmcp install /home/rj/Code/mcp-generated/whatsapp_server.py
  2. Configure Environment Variables:
    • Sign up for a Twilio account if you don't have one.
    • Get your Account SID and Auth Token from the Twilio Console.
    • Set up the Twilio Sandbox for WhatsApp or configure a dedicated Twilio WhatsApp number.
    • Create a file named .env in the project root directory (/home/rj/Code/mcp-generated/).
    • Add your Twilio credentials and WhatsApp number to the .env file: 🔑
      # /home/rj/Code/mcp-generated/.env TWILIO_ACCOUNT_SID=ACxxxxxxxxxxxxxxxxxxxxxxxxxxxxx TWILIO_AUTH_TOKEN=your_auth_token_here TWILIO_WHATSAPP_NUMBER=+14155238886 # Use your Twilio WhatsApp number (Sandbox or purchased)
    • Important: Replace the placeholder values with your actual credentials and number. Ensure the TWILIO_WHATSAPP_NUMBER starts with a + and includes the country code (E.164 format). The whatsapp_server.py script will automatically add the + if it's missing from the .env file. ❗
    • Trial Account Note: If you are using a Twilio trial account, the TWILIO_WHATSAPP_NUMBER will likely be the Twilio Sandbox number (+14155238886). You must enroll any recipient (to_number) phone numbers in your Twilio Sandbox via the Twilio console for messages to be delivered successfully. Sending to non-enrolled numbers requires upgrading your Twilio account.

Usage

1. Running the FastMCP Server

To make the send_whatsapp tool available for remote calls (e.g., from an AI model integrated with FastMCP):

python /home/rj/Code/mcp-generated/whatsapp_server.py

The server will start and print the Twilio number it's configured to use. It will listen for incoming requests (by default via stdio, but FastMCP supports other transports). A FastMCP client can then call the send_whatsapp tool with to_number (including the whatsapp: prefix, e.g., whatsapp:+15551234567) and message arguments.

Example Interaction (Conceptual):

A client (like an AI 🤖) might send a request like this (format depends on the transport):

{ "tool_name": "send_whatsapp", "arguments": { "to_number": "whatsapp:+15551234567", "message": "Hello from the FastMCP server!" } }

The server will process this, call the Twilio API, and return a confirmation or error message.

2. Running the Test Script

The /home/rj/Code/mcp-generated/twilio_test.py script provides a way to directly test sending a message using your Twilio credentials without the FastMCP server. 🛠️

  • Modify the script: Ensure the to= number in /home/rj/Code/mcp-generated/twilio_test.py is a WhatsApp number linked to your Twilio Sandbox (if using the Sandbox) or any valid WhatsApp number (if using a purchased Twilio number). The from_ number should typically be your Twilio Sandbox number (whatsapp:+14155238886) or your purchased Twilio WhatsApp number.
  • Run the script:
    python /home/rj/Code/mcp-generated/twilio_test.py
    This will attempt to send a hardcoded message ("Is this working?") from the specified Twilio number to the specified recipient.
-
security - not tested
-
license - not tested
-
quality - not tested

remote-capable server

The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.

A simple server that enables AI models to send WhatsApp messages via the Twilio API by exposing a send_whatsapp tool through the FastMCP framework.

  1. Features
    1. Setup
      1. Running the Test Script

    MCP directory API

    We provide all the information about MCP servers via our MCP API.

    curl -X GET 'https://glama.ai/api/mcp/v1/servers/wubbyweb/mcp-generated'

    If you have feedback or need assistance with the MCP directory API, please join our Discord server