https://github.com/pcastiglione99/ragify-search

RAGify is designed to enhance search capabilities using Retrieval-Augmented Generation (RAG). By combining traditional web search with AI-driven contextual understanding, RAGify retrieves relevant information from the web and generates concise, human-readable summaries.
https://github.com/pcastiglione99/ragify-search

langchain llm ollama python rag streamlit

Last synced: about 23 hours ago
JSON representation

RAGify is designed to enhance search capabilities using Retrieval-Augmented Generation (RAG). By combining traditional web search with AI-driven contextual understanding, RAGify retrieves relevant information from the web and generates concise, human-readable summaries.

• Host: GitHub
• URL: https://github.com/pcastiglione99/ragify-search
• Owner: pcastiglione99
• License: mit
• Created: 2024-12-13T17:10:42.000Z (about 2 months ago)
• Default Branch: main
• Last Pushed: 2025-01-13T07:18:39.000Z (18 days ago)
• Last Synced: 2025-01-13T08:26:35.829Z (18 days ago)
• Topics: langchain, llm, ollama, python, rag, streamlit
• Language: Python
• Homepage:
• Size: 234 KB
• Stars: 5
• Watchers: 1
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# RAGify Search - Intelligent Assistant with Web Search RAG

![RAGify](./RAGify.gif)

## Description

RAGify Search is an AI-powered intelligent assistant built with Streamlit, designed to answer user queries by integrating real-time web search. It uses a Retrieval-Augmented Generation (RAG) approach to provide accurate, concise, and context-aware responses

---

## Key Features

- **Real-Time Web Search**: Fetch relevant web pages for user queries.

- **Document Processing**: Split, embed, and index documents for similarity-based searches.

- **Prompt Engineering**: Generate optimized prompts for the LLM model.

- **Streamlit Interface**: Simple and intuitive chatbot-like user experience.

- **Temporary File Management**: Automatic cleanup of downloaded content to save space.

- **Local LLM Support**: Utilizes Ollama to run a local LLM, ensuring data privacy and high performance.

---

## Directory Structure

- **`app.py`**: Front-end application powered by Streamlit.

- **`config.py`**: Centralized configuration for chunk sizes, model names, and directories.

- **`db_operations.py`**: Embeds and indexes document chunks for search.

- **`extract_queries.py`**: Generates optimized queries to be searched.

- **`web_scraper.py`**: Manages web scraping and temporary file handling.

- **`prompt_generator.py`**: Prepares structured prompts for context-aware answers.

---

## Installation

1. **Clone the Repository**:

   ```bash

   git clone https://github.com/pcastiglione99/RAGify-Search.git

   cd RAGify-Search

   ```

2. **Install Dependencies**:

   ```bash

   pip install -r requirements.txt

   ```

3. **Run the Application**:

   ```bash

   streamlit run ./src/app.py

   ```

---

## Usage

1. Start the Streamlit app and interact with the chatbot.

2. Enter your query in the chat interface.

3. The assistant fetches web pages, processes content, and generates a response.