Google Vision

About

Unlock advanced document processing for large language models with this Model Context Protocol (MCP) server, powered by Google Cloud Vision API. Designed to overcome the inherent limitations of LLMs in handling non-textual content, this tool enables them to effortlessly read scanned PDFs, decipher handwritten notes, extract text from images, and process documents regardless of orientation. Initially developed for complex legal document analysis by Kohen Avocats, it provides a robust, standardized interface to bring powerful OCR capabilities directly to any LLM.

Key Features

• High-precision OCR for printed and handwritten text
• Multi-page PDF support (up to 2000 pages)
• Automatic document orientation detection
• Access to 9 Google Vision API features (text, document, labels, faces, objects, logos, landmarks, web, safe_search)
• Intelligent retry mechanism with exponential backoff
• 0 GitHub stars

Use Cases

• Populating Retrieval Augmented Generation (RAG) systems with extracted text
• Legal firms processing scanned documents and handwritten correspondences
• Mass OCR processing for multi-page PDFs with automatic handling