Do you have enough docs in store?

UnlimitedDocWorks

UnlimitedDocWorks is a Retrieval-Augmented Generation (RAG) system designed to process both unstructured documents and structured CSV data. It enables intelligent querying across diverse data sources, using state-of-the-art language models and orchestration tools.

The system utilizes Claude 3.5 Sonnet for natural language understanding and LangChain for integrating data retrieval, parsing, and response generation workflows.

Features

• Query over documents and CSVs with source-aware responses
• Support for both unstructured (e.g., PDFs, text files) and structured (e.g., tabular CSV) data
• Backed by Claude 3.5 Sonnet and LangChain for accurate, context-rich results
• Modular and extensible full-stack architecture

Setup Instructions

Follow the steps below to set up the project locally.

1. Clone the Repository

git clone <your-repo-url>
cd UnlimitedDocWorks

2. Download Required Files

Access the required assets from the following Google Drive link and download both files:

Google Drive – Project Files

3. Place the SQLite File and the archive folder

Move the downloaded .sqlite file into the following directory:

UnlimitedDocWorks/sqlite_store/

And the downloaded archive folder into :

UnlimitedDocWorks/

4. Install Dependencies

Install the required Node.js and Python dependencies from the project root:

npm install
pip install -r requirements.txt
Put api key in main_pipe in UnlimitedDocWorks folder

Running the Application

The application requires both the backend and frontend services to be running in separate terminals.

Terminal 1: Start the Backend

cd src
python backend.py

Terminal 2: Start the Frontend

# From the project root
npm run dev

License

This project is provided as-is under your chosen license. Ensure any third-party dependencies used comply with your intended usage and distribution.