Top all-time · #24406
Zach Ahn
Personal blog · zachahn.com/postsprogrammingai
Zach Ahn's personal blog at zachahn.com covers programming and ai. Since 2020, 3 different Hacker News readers have linked it 4 times, most often “How to fix Claude 5's token vomit”. It is still writing: the last post was on 2026-08-20, and you can follow it by feed.
Too few sampled comments (2) to say how often HN recommends it.?
Why HN links it
HN readers link this post about Claude 5 emitting excess tokens, such as hook instructions repeated verbatim, and about using a separate LLM to clean up that output. Commenters cite it as a source for their own complaint.
Summarised by Claude from a sample of 2 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Most-linked posts?
- How to fix Claude 5's token vomit2 people · 3 mentions · first linked 2026-08-20
What Hacker News says?
I totally agree that hooks help to shovel our instructions through to Claude, but it's so dumb we have to waste tons of tokens (repeated verbatim, over and over) (that we pay for), just to have it ignore the instructions anyway. I wrote a little bit about it on my blog post. It's a waste of money and compute. https://zachahn.com/posts/1787191554
Comment linking /posts/1787191554
on “Clean up Claude 5's token vomit with a separate LLM”Submitted How to fix Claude 5's token vomit as a story · 3 points
HN discussionAhh! Author here. Thanks for posting! In case anyone wants to share in my frustration, I also wrote a blog post about this. Blog post: https://zachahn.com/posts/1787191554 The prompt I use to tell the LLM what to fix: https://github.com/zachahn/vomit/blob/main/internal/config/s... Wasn't received too well on Lobsters haha, wrote a small extra blurb about it there: https://lobste.rs/s/juekuk/how_fix_claude_5_s_token_…
Comment linking /posts/1787191554
on “Clean up Claude 5's token vomit with a separate LLM”Submitted Deobfuscated Topological Sort as a story · 1 points
HN discussion