Sherlock Bot - WhatsApp CV Review Service

A serverless WhatsApp-based CV review bot using Firebase Cloud Functions, Twilio's WhatsApp API, and Stripe for payments.

Overview

Sherlock Bot allows users to:

• Upload their CV (PDF/DOCX) via WhatsApp
• Choose between Basic (free) or Advanced (paid) reviews
• Receive insights directly on WhatsApp
• Process payments seamlessly with Stripe
• Get detailed CV improvement recommendations

Architecture

The project uses a serverless architecture built on Firebase:

• Firebase Cloud Functions: Handles webhooks, processing, and business logic
• Firebase Firestore: Stores user sessions and review results
• Firebase Storage: Securely stores CV files temporarily
• Twilio WhatsApp API: Provides the messaging interface
• Stripe: Handles payment processing

Repository Structure

sherlock-bot/
│
├── functions/                      # Firebase Cloud Functions
│   ├── index.js                    # Main entry point
│   ├── src/
│   │   ├── config/                 # Configuration files
│   │   ├── controllers/            # Request handlers
│   │   ├── services/               # Business logic
│   │   ├── utils/                  # Utility functions
│   │   └── models/                 # Data models
│   │
│   ├── package.json                # Node.js dependencies
│   └── .env.example                # Environment variables template
│
├── firebase.json                   # Firebase configuration
├── firestore.rules                 # Security rules for Firestore
├── firestore.indexes.json          # Indexes for Firestore queries
├── storage.rules                   # Security rules for Firebase Storage
│
├── scripts/                        # Deployment scripts
│
├── .github/                        # GitHub Actions workflows
│   └── workflows/
│       └── deploy.yml              # CI/CD pipeline
│
├── docs/                           # Documentation
│
└── README.md                       # Project overview

Setup Guide

Prerequisites

• Firebase account with Blaze (pay-as-you-go) plan
• Twilio account with WhatsApp API access
• Stripe account for payment processing
• Node.js (v16 or higher)
• Firebase CLI

Installation Steps

1. Clone the repository

   git clone https://github.com/yourusername/sherlock-bot.git
   cd sherlock-bot

2. Create a Firebase project

   firebase login
   firebase projects:create sherlock-bot-xyz
   firebase use sherlock-bot-xyz

3. Install dependencies

   cd functions
   npm install

4. Configure environment variables

   cp .env.example .env
   # Edit .env with your API keys and configuration

5. Deploy to Firebase

   cd ..
   ./scripts/deploy.sh

6. Configure Twilio webhook

   • In your Twilio console, set up the WhatsApp Sandbox
   • Configure the webhook URL to point to your Cloud Function

   https://us-central1-sherlock-bot-xyz.cloudfunctions.net/api/webhook
   

7. Configure Stripe webhook

   • In your Stripe dashboard, set up a webhook
   • Configure the endpoint to point to your Cloud Function

   https://us-central1-sherlock-bot-xyz.cloudfunctions.net/api/payment-webhook
   

   • Select the checkout.session.completed event

Usage

Once set up, users can interact with the bot by:

1. Sending "Hi" or "Review CV" to your Twilio WhatsApp number
2. Choosing between Basic or Advanced review
3. Uploading their CV document
4. Receiving review feedback directly in WhatsApp

Development

Local Development

1. Start the Firebase emulators:

   firebase emulators:start

2. Set up Twilio to work with your local environment:

   • Use ngrok to expose your local server

   ngrok http 5001

   • Update your Twilio webhook URL to point to the ngrok URL

Adding New Features

• Create new service files in functions/src/services/
• Extend controllers in functions/src/controllers/
• Update security rules as needed in firestore.rules and storage.rules

Maintenance

File Cleanup

Files are automatically deleted after 24 hours using a scheduled Cloud Function. This ensures data privacy and compliance.

Monitoring

Use Firebase Console to monitor:

• Function invocations and errors
• Storage usage
• Firestore operations

License

MIT License - See LICENSE for details.

Contact

For questions or support, please contact your-email@example.com