API usage

[alfredit]

hi, really love the miniflux app, i want to ditch my own rss reader to this one (so i loose less time in coding and gain in reading articles.).
It lacks the one ai fuction i really love so i want to code this on my server with the API url and some curls.
I want to implement one thing :

• The the articles titles / unique id and ask gemini to drop those about commercials/spam links. I know using a llm for this is a lot since there is blocking rules but I've seen web sites are getting very clever to not being filtered and llm works great as a antispam filter !
  I've seen get entry and update entry but nothing to remove an entry. Is the a way to apply a "spam" group or delete it? I can rename it SPAM- if not possible.

Thanks ,

Edgar

[ckuethe]

What about marking it as read? I just put together a python script that does some additional analysis to decide what's boring (and marks it as read) or really interesting (and adds a star).

[alfredit]

Mark as read is a good idea. Do you share those py scripts so i can try and adapt them with my ideas? Thanks

[ckuethe]

I've extracted the logic. This should get you to a place where you can take an article (a dict), look at everything about it including the content, and make a decision on whether you want to call update_entries(list_of_article_ids, "read").

https://gist.github.com/ckuethe/231a529519d829f580679ddcc4eb04e9