Show HN: Hacker News with AI stories filtered out

hcker.news

52 points by postalcoder 1 day ago

Hi folks,

I've noticed that a lot of people here seem exhausted by the amount of AI news on the front page. I shared hcker.news here a year ago and it has since gained a ton of filtering features, including a dedicated AI filter, so I figured I should tell more people about it.

The filter works in three passes:

  1. Known AI-related keywords and domains are filtered automatically.
  2. An agent reviews the remaining articles and removes those it identifies as AI-related.
  3. I make the call on uncertain cases.

You can also exclude GitHub repos posted to HN that show signs of AI authorship. The filter checks:

  - Commit messages for agent attribution
  - Contributors for known coding agents
  - Repository files for agent instructions and configs

No filter will be perfect, but this is as good as it gets. lmk what you think. Cheers.

More links:

Original Show HN: https://news.ycombinator.com/item?id=44075353

If you only want AI stories: https://hcker.news/?view=frontpage&ai=include

RSS: https://hcker.news/feeds/atom?period=day&ai=exclude

A github-only non-AI feed: https://hcker.news/?ai=exclude&include_domains=github.com

rdmuser 14 hours ago

Hey thanks for adding customizable rss feeds! Super handy feature

Is there any chance the feed items could also have a direct link to the hn post? Right now I'm trying them, clicking through to your site then clicking to go to hn when I mostly just want the direct link to the hn discussion page so it's a bit clunky. I can solve this on my end easily with a redirect but I figured I'd mention it in case it's useful feedback.

coldpie 1 day ago

I also built a browser extension to do this client-side. It just auto-clicks the Hide links for any story with an AI-related term. Source & Firefox add-on link here: https://github.com/ColdPie1/enough-ai-on-hn

  • postalcoder 1 day ago

    There are a lot of ways of tackling this, but the only way to truly filter everything out is to update the list constantly.

    My filter currently has 146 keyword filters and 101 domain filters. Even with those, I've had to filter almost 400 exact urls on top of that in the past two weeks!

    • coldpie 1 day ago

      Definitely a lot of approaches. I like that mine is dead simple & runs in the browser; but an advantage to a deeper inspection like yours is it will catch those endless AI thinkpieces that don't obviously stick an AI term in the title. Pros & cons!

k310 1 day ago

I don't normally read the front page, more /newest

Any chance of implementing that?

Your site works with my Firefox auto-pager, which is really nice for archiving posts for later search.

  • postalcoder 1 day ago

    Hmm, right now I only process a subset of all submissions to keep things lean, so this is the closest that the site can do to /newest right now: https://hcker.news/?filter=top200&ai=exclude

    Feel free to email me.

    • k310 1 day ago

      No problem, and thanks. I understand the size of the problem, and why the above link is "most upvoted"

      By archiving all, I can easily filter titles as text, if I really wanted to do so, but it's not all that important for me. grep -v works!

      Again, thanks for the effort.

LunicLynx 1 day ago

Awesome! Finally the hacker news I originally came for