‹ BackHN Continuity

Thread

Agents don't need memory, they need documentation

365 points · 261 comments · kmeh

  1. spike021 · · focus · HN ↗
    I think whichever one is used, there needs to be a way to enforce what's written.

    If I say "use jq instead of writing a python script to parse json" it should never write adhoc python scripts to parse json. Yet that constantly happens to me anyway.

    1. jkhdigital · · focus · HN ↗
      Treat it like any other software system: rules that must not be violated are enforced by static type-checking or a trusted runtime monitor. There’s no other option.
      1. koolba · · focus · HN ↗
        Except you can’t do that unless the runtime itself can reason about what’s being executed.

        Otherwise you can get a python one liner that execs a different script engine.

        1. devmor · · focus · HN ↗
          You absolutely can, that’s what your harness is for. You don’t need your environment to “reason” about things when deterministic tools exist - You have a really fancy hammer, but that doesn’t make everything a nail.
          1. altmanaltman · · focus · HN ↗
            Yet the hammer seller continues to scream everything is a nail and their hammer will replace your entire job eventually. So are you telling me the hammer seller is lying or am I the one using it wrong?
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.