Document RAG Pipeline Skill

Overview

This skill creates a complete Retrieval-Augmented Generation (RAG) system from a folder of documents. It handles:

• Regular PDF text extraction
• OCR for scanned/image-based PDFs
• DRM-protected file detection
• Text chunking with overlap
• Vector embedding generation
• SQLite storage with full-text search
• Semantic similarity search

Quick Start

# Install dependencies
pip install PyMuPDF pytesseract Pillow sentence-transformers numpy tqdm