• rhamph@awful.systems
    link
    fedilink
    English
    arrow-up
    7
    ·
    17 小時前

    AGENTS.md serves an invaluable purpose in prominently tagging slopware. A mark the promptfondlers can’t wait to bolt on personally!

    If its content does nothing, does that mean that ‘poisoned’ contents are also nearly ineffective? Then it can be an even more straightforward slop signal.

  • G_M0N3Y_2503@lemmy.zip
    link
    fedilink
    English
    arrow-up
    2
    ·
    17 小時前

    I didn’t think these files were supposed to do anything more than this. Its supposed to provide a loose framework of the project, its supposed to trigger the agent to use specific tool more often.

    TL:DR it’s a hint that apparently costs 20% more.

    • Architeuthis@awful.systems
      link
      fedilink
      English
      arrow-up
      2
      ·
      7 小時前

      I think it also depends on how all in you are on vibecoding. Like if you just prompt it and then occasionally check if the token carnage produced anything usable, then having an agents.md is probably just best practices cargo culting. But in other cases, having a Here’s How Thing Are Done In This Repo, You Fucker markdown file should make going through the more slop-happy merge requests a little less soul crushing.

      The +20% additional token churn is surprising, I would have though the markdown files get cached pretty early down the line, unless the markdown file is really big compared to the chatbot output.

  • diz@awful.systems
    link
    fedilink
    English
    arrow-up
    15
    ·
    edit-2
    1 天前

    Fucking knew it. I follow some idiot slopper thread and they are all always going on about their prompting and absolutely none of them ever even considered measuring the effects.

    Having actually done optimizations including professionally once upon a time, its so ridiculous to see people thinking they optimize a stochastic process [when they don’t even formally measure anything]. It is just gambling.

  • Soyweiser@awful.systems
    link
    fedilink
    English
    arrow-up
    3
    ·
    21 小時前

    On the bright side, you can reduce your token usage by 20% with one simple trick.

    Or reduce it by almost 100% by using the ‘rm -rf’ command.

  • Architeuthis@awful.systems
    link
    fedilink
    English
    arrow-up
    12
    ·
    1 天前

    I mean it’s great if you want to automate some specific code generation in the most inefficient way possible while posing to the bosses as a forward thinking AI-first developer.

    The closest anyone can come to a use for the AGENTS.md file is so you clarify your own thinking on how the project works

    That just means that you are now supposed to write skill files instead of proper documentation to make your project even more inscrutable to anyone but a chatbots with a bottomless token budget.

  • Feyd@programming.dev
    link
    fedilink
    English
    arrow-up
    11
    ·
    2 天前

    Just the other day I was asking a coworker who wanted to add these to my repos if he’d compared if these actually do anything and of course he just viewed it as “something you’re supposed to do”.

    I’ve thought all along they look like marketing more than anything else.