Upload an excel spreadsheet. A pipeline labels each cell using the name the author already gave it. It makes it easier to understand and my thesis is that it will also help AI models reason faster and better. I have spent 20 years and 10,000 hours building and reviewing spreadsheets, in consulting and investing. The Trace tool would have made me catch the time I accidentally summed a date cell into a revenue total and would have saved me 1,000 hours of A1 hell. Labeling is currently a heuristic pipeline that detects common label patterns (year strips, text strips next to data, etc) and then a "walker" walks each cell till it finds its label. After this stage, remaining issues are sent to an AI (this takes a while on large sheets). Note -- still many edge cases but it works decently for well constructed spreadsheets - e.g. it still gets confused when authors put a label to the right of data cells, with sparse tables mixing text data and numerical data and with text and date formulas serving as labels, etc. Ts/Py stack. To see a cell's labels, upload a file, click on any cell and click the trace button. Pls give it a try.