This shit is more alpha than Vin Diesel, don't even use it.
The longterm goal is to provide to the CLI a .json config of desired html attribute/values to read, an output csv file path, and either a sitemap, input files, or json response and it will write the found values to csv.
Currently it is hardwired to find a couple meta tags from HubSpot from n files and writes them to a CSV.