About
Chomper is a powerful Model Context Protocol (MCP) server designed to efficiently process and understand a wide array of document formats for AI systems, particularly Claude. It excels at 'chomping through' over 36 file types, from PDFs and Office documents to code and emails, extracting rich text, images, and comprehensive metadata. With innovative features like Token-Optimized Object Notation (TOON) for approximately 40% token reduction and embedding-based semantic chunking, Chomper ensures AI models receive highly relevant and cost-effective document context, making it ideal for robust Retrieval Augmented Generation (RAG) pipelines and sophisticated document analysis.
