• I’m particularly interested in the disable-model-invocation: true safety on /exec and /cleanup. It addresses the biggest hurdle for AI in infra: the fear of an agent hallucinating a delete or exec command in the wrong context.
    • In addition the skills teach the agent about which kubectl commands are mutating so it doesn’t treat them and normal Bash() commands and asks you for permission first.
  • I wonder whether the future of software is like this project, where computer capabilities come in the form of skills that you can (purchase?/rent? to) download and use.
    • I'm reminded of Factorio.

      At first, it's really fun to make your own bases. Eventually, it seems you either lose interest, or you get impatient and copy in others' blueprints. A blueprint is just a plain-ish text representation of a part of a factory, and late game (or creative mode), the robots just build it. All you do is copy and paste it, not much creativity required.

      • Where in Computer Science or business process engineering is the creativity going to go?
    • What should we call it? Software as a Service?
      • Shareware suggests itself.
  • [dead]
  • [flagged]
  • [flagged]