Show HN: Aharness – Enforce coding-agent workflows as state machines on Codex

1 comments

• rishabhpoddar 1 hour ago
  As models get better and better at following instructions, don't you think this will become lesser and lesser useful?
  [-]
  • alfredvc 1 hour ago
    The better the models, the fewer guardrails they will need. However, the more you know about the process the more useful the guardrails are. Guardrails may become less useful over time, but you pay for it in uncertainty and likely token usage.