WhatsApp Chatbot Connector

This project is a WhatsApp Chatbot Connector backend built using Express.js. It is designed to integrate with the WhatsApp Business API and supports various AI platforms such as Dify and Rasa.

Here is the diagram to understand the flow:

Features

• Webhook endpoint to receive messages from WhatsApp
• Integration with Dify and Rasa for natural language processing and response generation
• Displaying interactive whatsapp message like List, Reply Buttons, and Flow

Prerequisites

Ensure you have the following installed:

Node.js (v18.x or later) npm (v10.x or later)

Getting Started

Quick Start

1. Clone the Repository

git clone https://github.com/hyperjumptech/whatsapp-chatbot-connector.git
cd whatsapp-chatbot-connector

2. Install Dependencies

npm install

3. Set Environment Variables

cp .env.example .env

then fill in the appropriate values in the .env file.

Choose to connect to which platform by filling the CONNECTION_PLATFORM variable in the .env file.

CONNECTION_PLATFORM=dify

or

CONNECTION_PLATFORM=rasa

4. Running the Server

npm run dev

5. Set callback URL in WhatsApp Business API

In the WhatsApp Business API, go to the Configuration menu and fine the Callback URL. Detail on the picture below, poin number 7

Note that the url must use https

So if you are running on local, you can use ngrok to get a public URL with https for your local server.

6. Send text message to the WhatsApp Business API

Now just use your WhatsApp app to send a text message to the WhatsApp Business number and you will see the response from the chatbot connected.

Optional Development Dependencies

To start up the optional development dependecies like Redis, you can use Docker .

Go to folder .dev then

docker compose up -d

Environment Variables

Variable Name	Description	Example
NODE_ENV	Environment variable to set the node environment.	development
WEBHOOK_VERIFY_TOKEN	Webhook verification token. The value should be the same as the one you set in the WhatsApp Business API. Detail in picture below, poin number 3
GRAPH_API_TOKEN	Graph API token. The value should be the same as the one you set in the WhatsApp Business API. Detail in picture below, poin number 5	abacdefghijk
BUSINESS_PHONE_NUMBER_ID	Business phone number ID. The value should be obtained from WhatsApp Business API. Detail in picture below, poin number 6	12345678912323
DIFY_BASE_URL	Dify base URL.	https://api.dify.ai/v1
DIFY_API_KEY	Dify API key.	app-Jdasdsdsd98n98787y
RASA_BASE_URL	Rasa base URL.	http://localhost:5005/webhooks/rest/webhook
CONNECTION_PLATFORM	Platform to connect to. value should be dify or rasa	dify
SESSION_DATABASE	Database to store session. value should be in-memory or redis	in-memory
REDIS_URL	Redis URL. Required if SESSION_DATABASE is set to redis	redis://localhost:6379

WhatsApp Business API Configuration

Interactive Message Configuration

Flow

To create a Flow, you must setup many things like in the docs. I will mention over-simplified version for easy starting point:

1. Create a Flow
2. Setup server endpoint
3. Setup Dify to trigger sending HTTP Request with this JSON format. Here is the example request body:

{
  "recipient_type": "individual",
  "messaging_product": "whatsapp",
  "to": "{{#sys.user_id#}}",
  "type": "interactive",
  "interactive": {
    "type": "flow",
    "header": {
      "type": "text",
      "text": "Event Registration"
    },
    "body": {
      "text": "Click button below to authenticate yourself"
    },
    "action": {
      "name": "flow",
      "parameters": {
        "mode": "draft", // delete this field for production use (when your flow is published)
        "flow_message_version": "3",
        "flow_token": "auth_flow-{{#sys.user_id#}}.",
        "flow_id": "1234567812345678",
        "flow_cta": "Start",
        "flow_action": "data_exchange"
      }
    }
  }
}

4. It will send CTA button to trigger the Flow
5. Fill in, complete, and submit the form
6. Sent status will be displayed

Deployment

Vercel

You can deploy this app to any server that runs Node.js. The easiest one is to use Vercel. Just clone this repo and connect it from Vercel Dashboard then you are good to go.

Docker

Use Dockerfile and docker-compose.yml file in the root folder.

1. Build docker image

docker build -t wa-connector

2. Run the docker

docker compose up -d

License

This project is licensed under the MIT License. See the LICENSE file for details.