patent-pdf-to-md

Convert Chinese patent PDFs/DOCX/DOCs to structured Markdown. Supports patent publications (A/B/U) and office action documents (审查意见通知书/驳回决定/复审决定书/无效宣告请求审查决定书). Uses MinerU for high-quality OCR text extraction (falls back to Tesseract), Python scripts for structural parsing and image extraction. Invoke when user needs to process Chinese patent PDFs, convert…

Independent GitHub snapshot · Automated classification · Original documentation governs