Overview

Use the WhatsApp Business API and Brainfish API together to automatically respond to incoming WhatsApp messages with context-sensitive answers.

When to Use This Guide

• Use this guide if you want business automation: customers message a WhatsApp Business number and a server replies using Brainfish.
• If the goal is setting up a personal WhatsApp account, this guide does not apply (install the WhatsApp app and verify a phone number instead).

High-Level Setup (Quick Summary)

1. Set up a small web service environment (example uses Node.js) and install dependencies (e.g., express, axios).
2. Generate a Brainfish API token.
3. Set up WhatsApp Business API in Meta (Business account, phone number, templates as needed).
4. Create a webhook server that:
   • Receives inbound WhatsApp webhook events
   • Extracts the user message
   • Calls the Brainfish API
   • Sends Brainfish’s response back via the WhatsApp Business API
5. Configure the webhook callback URL in the Meta Developer Dashboard and complete webhook verification.
6. Test end-to-end by messaging the WhatsApp number.

---

This guide will teach you to create a system that automatically responds to WhatsApp messages using the Brainfish API. We’ll cover setting up a server, configuring WhatsApp Business API, and writing code to bridge Brainfish and WhatsApp.

---

Prerequisites

• Brainfish API Access: You’ll need an API token from Brainfish.
• WhatsApp Business API Account: Set up via Meta for business use cases.
• Node.js & npm: Install these to run a server.

---

Step 1: Setting Up Your Development Environment

1. Install Node.js: Download and install Node.js, which includes npm (Node Package Manager).
2. Set Up a New Project:

   • Open a terminal.

   • Create a project directory and initialize it with npm:

     bash
     

     Copy code

     mkdir whatsapp-brainfish-integration
     cd whatsapp-brainfish-integration
     npm init -y
     
     

3. Install Required Packages:

   • We’ll use express for handling HTTP requests and axios for making API calls.

     bash
     

     Copy code

     npm install express axios
     
     

---

Step 2: Generate a Brainfish API Token

1. Go to Brainfish Settings → Advanced → API Tokens: Brainfish API Token Page.
2. Click "Create New Token", name it (e.g., "WhatsApp Integration"), and copy the token.

⚠️ Important: Keep this token safe—it should be used only on secure servers.

---

Step 3: Set Up WhatsApp Business API

To send automated messages on WhatsApp, you need a WhatsApp Business API account.

1. Sign up on Meta for WhatsApp Business API: Visit WhatsApp Business API Getting Started.
2. Get a Phone Number and Approve Message Templates:
   • Register a phone number dedicated to your WhatsApp API.
   • Submit and approve message templates (e.g., "Hello! Thank you for reaching out. We'll get back shortly.").
3. Set Up Webhook:
   • In your Meta Developer Dashboard, under Webhooks, configure the webhook URL (we’ll create it next) where incoming messages will be sent.

---

Step 4: Create the Server to Handle WhatsApp and Brainfish API Calls

Let’s create an Express server to handle incoming WhatsApp messages and make requests to the Brainfish API.

1. Create a new file server.js in your project directory.

2. Paste the following code:

   javascript
   

   Copy code

   const express = require('express');
   const axios = require('axios');
   const app = express();
   
   app.use(express.json());
   
   // Step 4.1: Webhook endpoint for WhatsApp
   app.post('/webhook', async (req, res) => {
     try {
       // Step 4.2: Extract incoming message and sender details
       const incomingMessage = req.body.text;
       const senderPhone = req.body.from;
   
       // Step 4.3: Query Brainfish API with incoming message
       const brainfishResponse = await axios.post(
         'https://app.brainfi.sh/api/articles.contextual.search',
         { query: incomingMessage },
         {
           headers: {
             'Authorization': 'Bearer YOUR_BRAINFISH_ACCESS_TOKEN',
             'Content-Type': 'application/json',
             'Accept': 'application/json'
           }
         }
       );
   
       const answerText = brainfishResponse.data.data.suggestedAnswerText;
   
       // Step 4.4: Send response back to WhatsApp
       await axios.post(
         `https://graph.facebook.com/v11.0/YOUR_PHONE_NUMBER_ID/messages`,
         {
           messaging_product: 'whatsapp',
           to: senderPhone,
           type: 'text',
           text: { body: answerText }
         },
         {
           headers: {
             'Authorization': 'Bearer YOUR_WHATSAPP_TOKEN',
             'Content-Type': 'application/json'
           }
         }
       );
   
       res.sendStatus(200); // Acknowledge receipt of the message
     } catch (error) {
       console.error('Error handling message:', error);
       res.sendStatus(500);
     }
   });
   
   const PORT = process.env.PORT || 3000;
   app.listen(PORT, () => {
     console.log(`Server is running on port ${PORT}`);
   });
   
   

   • Replace

     YOUR_BRAINFISH_ACCESS_TOKEN
     

     with the token from Brainfish.

   • Replace

     YOUR_PHONE_NUMBER_ID
     

     and

     YOUR_WHATSAPP_TOKEN
     

     with details from your WhatsApp API setup.

3. Start Your Server:

   • In the terminal, run:

     bash
     

     Copy code

     node server.js
     
     

   • Your server should now be running and listening for incoming WhatsApp messages.

---

Step 5: Configure Webhook on Meta Developer Dashboard

1. Go to your WhatsApp Webhooks section on the Meta Developer Dashboard.

2. Set the Callback URL to your server’s URL (e.g.,

   https://yourdomain.com/webhook
   

   ).

3. Verify the Webhook:

   • You may need to respond with a verification token; add this to your code if needed.

4. Save your changes.

---

Step 6: Test the Integration

1. Send a WhatsApp Message to your registered phone number (from Step 3).
2. Observe the Response:
   • Your server should receive the message, send it to Brainfish, and return an answer to WhatsApp.

---

Sample Interaction: Consumer Use Case

• Customer: “What is Generative Language Models?”
• Response: “A Generative Language Model (GLLM) is a sophisticated AI model designed to generate coherent and contextually relevant text...”

---

Additional Configuration Tips

• Logging: Use console logs or a logging service to monitor requests and troubleshoot errors.

• Error Handling: Update the

  catch
  

  block in

  server.js
  

  to handle specific errors (e.g., expired tokens).

• Testing on Localhost: Use tools like ngrok to expose your local server to the internet for testing webhooks.

---

Troubleshooting & Best Practices

• Token Management: Brainfish and WhatsApp tokens may expire. Keep track of expiry and renew tokens as needed.
• WhatsApp Message Limits: Check WhatsApp’s message rate limits to avoid restrictions.
• Compliance: Ensure responses comply with WhatsApp's guidelines.

---

Following these steps will set up an automatic response system for WhatsApp using Brainfish API from scratch, enhancing customer engagement with accurate, context-sensitive answers. For more support, reach out to support@brainfi.sh.