Every month I drop four PDF catalogues into a folder. Each page shows the rims in a fixed 4 × 3 grid, with the size written directly underneath every image. I’d like a Python script that picks up those PDFs automatically, pulls out every image together with its corresponding size, and then does three things: • Saves each image to a month-named sub-folder (e.g. “2024-07”) so the archive grows cleanly over time. • Builds an Excel file in the same run—two columns only: ImagePath (not the binary image) and RimSize. No embedding. • Lets me trigger the whole process with a single command or double-click, making it repeatable for future months without manual tweaks. Because the layout is identical on every page, the extraction can rely on position rules rather than complex OCR. I’m fine with whatever libraries you feel are best—PyMuPDF / pdfplumber for parsing and Pillow or OpenCV for image handling are all acceptable as long as the final package installs smoothly with a requirements.txt and runs on Windows. Deliverables • Clean, well-commented Python script • README showing setup, one-line execution and how to change the month folder name • Sample output (one Excel file + the images) generated from a single sample PDF so I can verify accuracy If the script reliably pulls each rim and its size into the Excel sheet and organises the images by month without embedding them, I’ll consider the job done.