isSpam: ML-Powered Email Spam Classifier Extension

isSpam is a full-stack project designed to classify an open email as spam or safe (ham) by analyzing its content using a trained Machine Learning model hosted as a private API service. The system is wrapped in a lightweight Chrome extension for real-time, in-browser classification.

✨ Project Components

Component	Technology	Role	Location
ML Model	Python, Scikit-learn (Logistic Regression)	Core logic: Predicts 0 (Ham) or 1 (Spam) based on text features.	api/models/
API Backend	Python, Flask, Gunicorn	Exposes the ML model as a secure, real-time HTTPS endpoint.	api/
Browser Extension	JavaScript (MV3)	UI wrapper: Extracts email text, calls the API, and displays the result as a banner.	extension/

🚀 Deployment Status

Service	Status	Endpoint / Location
API Service	LIVE	https://isspam-ys1h.onrender.com/predict
Extension	LIVE	https://chromewebstore.google.com/detail/isspam/giioaghhfkefecelfjjfldaagofjmfgg?pli=1
Google Workplace Marketplace add-on	Submitted for review	https://script.google.com/d/19xw2Urq0TLPHVZDgNNYuBJ-n7oZJFFIrRCxXd0HpnnnO7EYAPd3ZA2zE/edit?usp=sharing

🛠️ Setup and Installation

1. Browser Extension Installation

To use the extension, you must load it into your browser.

1. Package the Extension: Create a ZIP file of the contents of the extension/ directory. (The manifest.json file must be at the root of the ZIP).
2. Open Chrome Extensions: Navigate to chrome://extensions/.
3. Enable Developer Mode: Toggle the switch in the top right corner.
4. Load Unpacked: Click the "Load unpacked" button and select the extension/ folder.

2. Running the API Locally (Optional)

If you need to test or modify the Python backend locally, follow these steps:

1. Navigate to the api/ directory.
2. Install dependencies: Ensure all packages are installed, including the production server Gunicorn.

   pip install -r requirements.txt

3. Run the API Server: The server must be running before the extension can call it on localhost.

   gunicorn app:app --bind 0.0.0.0:5000

   Note: Your live extension is configured to call the Render URL, so you must change the API_ENDPOINT in extension/background.js to http://localhost:5000/predict to test locally.

💡 How to Use the Extension

1. Open your preferred web-based email client (e.g., Gmail or Outlook).
2. Open an individual email message so the full body content is visible.
3. Click the "Analyze Email" extension icon in your Chrome toolbar.
4. The extension extracts the text and injects a banner at the top of the email body showing the classification result:
   • Spam: ⚠️ ACTION REQUIRED: Classified as SPAM. Proceed with caution.
   • Ham: ✨ Looks Good: Message classified as SAFE.

⚙️ Project Architecture & Deployment

The project uses a clean separation of concerns for stability:

1. Extraction: The content.js file is injected upon a toolbar click (background.js) to perform DOM queries and extract the raw email body text.
2. Communication: The extracted text is passed to the background.js Service Worker.
3. Prediction: background.js makes a secure JSON POST request to the Render API (https://isspam-ys1h.onrender.com/predict).
4. Deployment: The Flask API is hosted by Gunicorn on Render, providing a stable, production-grade HTTPS endpoint.
5. Display: The API's integer prediction (0 or 1) is sent back to content.js, which dynamically creates or updates the result banner on the live webpage.