Enables AI agents to process files locally — OCR images, extract text from PDFs and DOCX, and describe images using local vision models, all without sending data to external services.