RAGsense: RAG powered AI Search Engine

RAGsense is a RAG powered AI search engine that allows you to ask questions about the content of any web page you are currently visiting. It combines a Chrome Extension with a Python backend powered by Retrieval-Augmented Generation (RAG). Built with 100 % open-source stack: LangChain, RAG, Hugging-Face embeddings + small LLM, FAISS vector store, and a tiny *Manifest-V3 Chrome extension

🚀 Features

• Local Intelligence: Uses google/flan-t5-small and sentence-transformers running entirely on your CPU. No data leaves your machine (unless you deploy the backend).
• Instant Answers: Ask questions about the current tab's content and get immediate, context-aware answers.
• RAG Pipeline: Built with LangChain and FAISS for efficient document retrieval and answering.
• Chrome Extension: Seamless integration with your browsing experience.

📋 Prerequisites

• Python 3.10 or higher
• Google Chrome (or Chromium-based browser)

📦 Installation

1. Backend Setup

The backend handles the scraping, embedding, and question-answering.

1. Clone the repository:

   git clone <repository-url>
   cd RAGsense

2. Create and activate a virtual environment:

   # Windows
   python -m venv venv
   .\venv\Scripts\activate
   
   # macOS/Linux
   python3 -m venv venv
   source venv/bin/activate

3. Install dependencies:

   pip install -r backend/requirements.txt

4. Start the server:

   python backend/main.py

   The server will start at http://0.0.0.0:8000.

2. Chrome Extension Setup

The extension allows you to interact with the backend from your browser.

1. Configure the Backend URL:

   • Open chrome-extension/popup.js.
   • Find the line:

     const BACKEND = "https://ai-search-engine-abc123-4567.app.github.dev";

   • Change it to your local server URL (or deployed URL):

     const BACKEND = "http://localhost:8000";

2. Load the Extension:

   • Open Chrome and navigate to chrome://extensions.
   • Enable Developer mode (toggle in the top right).
   • Click Load unpacked.
   • Select the chrome-extension folder from this project.

💡 Usage

1. Ensure the backend server is running (python backend/main.py).
2. Open any web page in Chrome (e.g., a news article or documentation).
3. Click the RAGsense extension icon in the toolbar.
4. Type your question in the input box (e.g., "What is the main summary of this article?").
5. Click Ask.
6. The AI will read the page content and provide an answer based on the text.

📁 Project Structure

RAGsense/
├── backend/
│   ├── main.py           # FastAPI application entry point
│   ├── rag_chain.py      # RAG logic (Embeddings, LLM, Vector Store)
│   └── requirements.txt  # Python dependencies
└── chrome-extension/
    ├── manifest.json     # Extension configuration
    ├── popup.html        # Extension UI
    ├── popup.js          # Extension logic
    ├── content.js        # Content script
    └── icons/            # Extension icons

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

AUTHOR: Phravins

RAGsense - Eclipse Public License - v 2.0 licenced