Ingests large PDF documents into a RAG pipeline with parallel parsing, S3 storage, and Postgres status tracking.
The AI RAG PDF Ingest skill provides a robust backend infrastructure for handling the document ingestion phase of Retrieval-Augmented Generation. It automates the complex workflow of uploading files up to 2GB to S3-compatible storage, extracting page-level text content using high-performance parallel parsing with the unpdf library, and maintaining processing states within a PostgreSQL database. This skill is essential for developers building AI applications that require reliable document processing, structured text extraction, and scalable storage management within a Next.js environment.
주요 기능
01Page-level text storage and metadata extraction
02Support for large PDF uploads up to 2GB to S3-compatible storage
03Parallel PDF parsing and text extraction using the unpdf library
04Pre-configured Next.js API routes for document lifecycle management
053 GitHub stars
06Real-time processing status tracking in PostgreSQL via Drizzle ORM
사용 사례
01Automating text extraction from large legal, medical, or technical manuals for LLM training
02Building 'Chat with your PDF' applications or AI-powered document search engines
03Creating centralized document processing hubs for enterprise RAG pipelines
