the-dataface/Newspaper-Scrapers — reverse-engineered prompt

Build me a simple Python tool that can search a bunch of major news sites for a keyword over a date range, then collect the article details into one clean dataset. I want it to work with outlets like the New York Times, Washington Post, Wall Street Journal, USA Today, CNN, Fox News, Politico, Slate, CNBC, Bloomberg, Time, Weekly Standard, LA Times, and Chicago Tribune.

The basic flow should be, give it a source name, search term, start date, and end date, have it find matching article links, pull the article metadata and text, and then save the results either to MongoDB or to a CSV file. The fields I care about are title, publish date, outlet, authors, feature image, article link, keywords, summary, full text, and raw HTML, plus any other metadata the parser can already extract.

If a site needs login access, like NYT or WSJ, make that optional through command line inputs. Please include an easy runner script so I can kick off scrapes quickly, and look up current docs online if you need to.