• Mikina@programming.dev
    0·
    21 hours ago

    Kind of have an urge now to post an issue that the nstall skill has installed a malware on my machine.

    It’s pretty plausible that could happen, and good luck debugging that.

  • Dima@feddit.uk
    0·
    2 days ago

    Unfortunately not the first time I’ve seen this sort of thing: Screenshot of install instructions with "Option A: Let an LLM do it"
    (I don’t use this project, just came across it)

    • AeonFelis@lemmy.world
      0·
      2 days ago

      I’ll give it a pass because it’s an OpenCode plugin, meaning the typical user is not tech-savvy enough to be able to edit a simple JSON configuration file.

      • renlok@lemmy.zarioth.com
        0·
        21 hours ago

        But surely it also means the plugin fully slopcoded and the maintainer has no idea what they are doing if they are incapable a writing instructions

  • FauxLiving@lemmy.world
    0·
    2 days ago

    You may as well just completely replace your package manager with this one simple script.

    #!/usr/bin/env python3
import sys

exec(TrustedAgent.request(f"Write a Python statement to install {sys.argv[1]}, make no mistakes"))
  • civ@lemmy.civl.ccEnglish
    0·
    2 days ago

    The logical next step is to ask a humanoid robot to interface with your machine to perform these steps for you

  • Evotech@lemmy.world
    0·
    2 days ago

    I regularly have ai agents install shit for me lol. I just point it at a repo and say «deploy this for me» (in sandboxed environments)

    But i question this approach very much.

    • NιƙƙιDιɱҽʂ@lemmy.world
      0·
      2 days ago

      I disagree with it but think it’s fine if that’s what you’re comfortable with for whatever you’re doing on your system, but for the repo maintainer to make that the de facto way to install your software is absolutely wild and speaks leagues of the quality of the software you’re about to install…

      • Evotech@lemmy.world
        0·
        2 days ago

        Yeah. That is bad.

        But, it does speak to the capabilities. I’m sure you could offload all sysadmin tasks to an LLM these days. Just have it ssh in and do whatever you wanna do.

        • littleomid@feddit.orgEnglish
          0·
          2 days ago

          I’m sure I could also allow a child to do the sysadmin tasks that I do, but that would be an extremely stupid thing to do. Our systems can’t break. AI breaks shit constantly.

        • pressanykeynow@lemmy.world
          0·
          2 days ago

          And as a bonus once in a while it will delete your database and backups so won’t ever need to do any sysadmin tasks at all. Optimisation!

  • Mikina@programming.dev
    0·
    2 days ago

    This will be a goldmine for malware groups that are poisoning AIs that use web search to get them to use their malware-ridden version of software and libraries, and I’m here for it with popcorn at the ready.

    I kinda hope this will get more common, because the only people affected kind of deserve it, plus it’s a great way how to filter sloprojects at a glance.

    • TheLegendaryAssholeOfJushinLiger@sh.itjust.works
      0·
      2 days ago

      I don’t think that poisoning really works. They have AIs judge the answers before they incorporate them into the new AI. They can tell the best topping for pizza isn’t actually mayonnaise and strawberries

  • Mikina@programming.dev
    0·
    2 days ago

    While I’m not a fan of most AI usages, this is the thing that infuriates me the most.

    I like writing scripts to automate parts of my job. I’ve had a few for things like build performance testing comparison and the like.

    All of it was replaced by skill.md, that does exactly the same, but burns like 3$ per run in tokens, and has also at least once generated hallucinated results, because it ran into an error, ran wrong builds, or in general fucked up in a way that was not easy to detect (and we did in fact not detect it until much later).

    But hey, at least my colleagues now don’t have to open the filthy commandline and write py perf-test.py main feature/branch to run the test, and can just talk about it to a clanker.