I’m looking to automate the way I extract questions from past JEE papers. The goal is a small, reliable tool that accepts whole papers in either PDF or Word format and returns a clean CSV file. Each row of that CSV must contain the full question text, every answer option in its own column, and a separate column for the correct answer. My papers include both multiple-choice and numerical-value questions, and many of them feature chemistry notations, complex mathematical expressions, and the occasional diagram. I need those elements preserved in a way that keeps the CSV usable—whether that means embedding LaTeX-style markup, inserting image links, or another technique you propose. I don’t mind if you build a command-line utility or a lightweight desktop app, as long as it runs on a Mac and processes a full paper in one go without manual intervention. Accuracy is more important than speed: the tool must recognise equations without mangling subscripts, superscripts, or symbols, and it should never drop a diagram. Deliverables: • Source code and a short README covering prerequisites, installation, and usage. • A compiled executable or script ready to run. • Ten sample runs proving that a mixed PDF/Word set becomes a well-formatted CSV exactly as described.