News for hackers. Stuff that matters.

A comment on Handbook.md shows that long policy documents do not reliably govern agents. Parent comment

Re:Handbook.md shows that long policy documents do not relia... by Analemma_ on Wednesday July 29, 2026 @12:23PM (#49099537)
In my own experience (both Claude and Gemini 1M), you start running into problems around 60-70%. A 25% ceiling sounds like total cargo culting to me.
[ Parent | on HN ]
Re:Handbook.md shows that long policy documents do not relia... by irthomasthomas on Wednesday July 29, 2026 @01:10PM (#49100208)
Benchmarks show a large drop in quality as context grows. Your opus will be acting like haiku above 300k tokens.
[ Parent | on HN ]