Basically I want a self-hosted version of Google News. If 10 feeds have near-identical coverage of the same story, I want just 1 item, with an option to see the other 9.

I went looking for something like this a couple years ago and found ~nothing. Checked again today and now there are Entirely Too Many Things.

Hoping to find something a bit more established than somebody’s abandoned weekend project. There seems to be an awful lot of that out there right now.

Looking for XML or similar output. Do not need or necessarily want any sort of front end.

  • bulwark@lemmy.world
    link
    fedilink
    English
    arrow-up
    20
    ·
    edit-2
    18 天前

    I’ve been really happy with Fresh-RSS. Someone else on here put me onto about 6 months ago and it’s changed how I consume news.

    I find the most time consuming part was/is curating my feed but with tools like RSS-Bridge I can really get fine grain control on what makes it through.

    On your LLM summarizer question, yes they have several plug-ins. I’ve recently started piping the whole feed into Gemini and telling it to pick the top 5 and summarize. But I do that in bash and emacs :

    • TheTonDog@feddit.uk
      link
      fedilink
      English
      arrow-up
      2
      ·
      18 天前

      This looks amazing , appreciate the steer on it.

      I don’t suppose you know if Kindle delivery is supported at all? I use my own dockerized fork of news-sender-kindle which works well in terms of folding everything down into a .epub, but the RSS control is a little fiddly, it doesn’t let me preview the articles and amending feeds involves SSHing into an txt file on my server (not ideal).

  • barkingspiders@infosec.pub
    link
    fedilink
    English
    arrow-up
    7
    ·
    18 天前

    I’ve been using miniflux in docker with a miniflux-ai container I found on GitHub, I seem to recall I had issues out of the box and had to tweak it but that was quite a while ago, can’t remember the details. I have it configured to generate summaries of a list of feeds in miniflux when it sees new items. Works pretty well, needs to be restarted occasionally but has been just fine for the amount of time it took to setup.

  • rumba@lemmy.zip
    link
    fedilink
    English
    arrow-up
    3
    ·
    18 天前

    Go check out network chucks review of N8N.

    He builds a small news aggregator with LLM that crap stuff into discord but you could easily put it anywhere.

  • InnerScientist@lemmy.world
    link
    fedilink
    English
    arrow-up
    3
    ·
    edit-2
    18 天前

    I’ve searched for something like this in the past and didn’t find anything.
    I’ve though of hacking together a few scripts or 8n8 to interface with freshrss to do this but it’s far down my list™

  • koala@programming.dev
    link
    fedilink
    English
    arrow-up
    1
    ·
    18 天前

    Remember that Google News has RSS feeds! They are very well hidden, but they are there.

    However, they are also a bit bad.

    I started https://github.com/las-noticias/news-rss to postprocess a bit Google News RSS feeds and also play with categorization. I found spaCy worked well to find “topics”, but unfortunately I lost steam.

  • artyom@piefed.social
    link
    fedilink
    English
    arrow-up
    1
    ·
    edit-2
    18 天前

    Honestly Brave News is the only solution I’ve found for that. Just go to Brave search and search for “news”. It will aggregate the top stories of the day, group the different sources, then provide an AI generated digest of the stories upon request.

    • ryanpdg1@lemmy.ca
      link
      fedilink
      English
      arrow-up
      1
      ·
      18 天前

      I’m with OP on this one. I enjoy reading quality content, not just the act of reading. I think the idea is that by using an LLM to help sort through the drivel you have more time to do the reading of the quality content.

      And really, I’m always suspicious of the why when an algorithm suggests me an article… Having something less tied directly to a companies bottom line might help me trust the content more.