So I was reading https://fedi.simonwillison.net/@simon blog post https://simonwillison.net/2023/May/18/cli-tools-for-llms/
He had a neat workflow to summarize web pages.
So I extended it, and used pup
and strip-tags
to get the frontpage links from the Drudge Retort (NOT DRUDGE REPORT), processed all the links through curl and use 4o
to summarize them. THen I summarized the summaries, and I get a quick hit read of what happened today. Kinda like WTFJHT.
# Create a file to store all summaries
echo -n "" > all_summaries.txt