According to Lumo, uploaded binary files are only read as a first snippled, and this is no different in the Lumo+ version. Hence, an Excel table or a pdf cannot be parsed or only the first few lines, rendering the function useless. Pls, parse full files. Output of Lumo:

File‑type handling: I can read the first few lines of a text‑based file (CSV, plain‑text, JSON) that you upload, but I cannot open binary formats (PDF, DOCX, XLSX) or run arbitrary parsing libraries. Your annual‑report PDFs and the Excel workbook itself are binary; I can see the file name and a tiny preview of a CSV‑style export, but I cannot extract tables or numbers from them.