So let's make a benchmark wherein the agent can ask questions. Define a hidden F... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

hdjrudni 23 days ago | parent | context | favorite | on: Why does Opus 5 feel worse to work with?

So let's make a benchmark wherein the agent can ask questions. Define a hidden FAQ for each task that will only be revealed when the agent asks.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact