I have a short list of specific websites that host online databases containing the numeric figures I need. The task is to pull those numbers—no text, images, or PDFs—into a single, tidy dataset. I’m only interested in the quantitative fields (prices, counts, ratings, percentages, etc.), and each site will be shared privately once we begin. The workflow I envision: programmatic access to each site (API if one exists, otherwise straightforward web-scraping with Python, Requests/BeautifulSoup, Selenium, or a comparable tool), parse the numeric fields, and output them in a consistent structure so I can reload the file whenever fresh numbers are posted. Deliverables • A clean CSV or XLSX containing all requested numeric fields, one row per record, with columns clearly labeled • The complete, well-commented script(s) used to collect and process the data • A brief README explaining how to rerun the collection and any dependencies I’ll validate the file by cross-checking a random sample on the live sites, so accuracy and repeatability are key.