Great site, triggered memories! haha.

To try to add something to this discussion -- I think that while I've seen these sort of loops less --- what I have seen is "overly helpful".

Models nowadays want to double-triple-quadruple check things. I'm being silly but it verges on "I have a working solution but let me write a variation in Rust to ensure a convergent solution and prove this works".

I've had to stop models nowadays mostly because they're being agonizingly pedantic in their validation. Opus is actually one of the most pedantic and "off track" here. But again, not in a bad way. I'm usually like "stop testing latency between 50 runs of this app... this is version one.. we're going to make a million more changes.. you're not buying us anything".

This is my recent experience as well. Models want to run linters, tests, etc. And that is all covered in GitHub actions. So I have been instructing agents to push a draft PR, then I validate the static checks pass and tell the agent if there are issues.

Agents and AI are getting expensive, it seems silly to waste tokens on static checks.

> triggered memories

Yeah of 10 minutes ago. It is shocking how long some seemingly simple things can take. I know there are some things I can do faster than the LLM and some things it can do faster than me. The amount of rambling BS is the exhausting part.

Claude models were always too eager and "overly helpful" for my taste. But it seems better models tend to be this way. GPT 6 and 5.6 are overly helpful too, but at least less than Fable. But I seem to be sticking to GPT 5.5 as this was a really focused model.

One thing it's missing: "smoking guns" and "smoke tests"

If you search my company's Slack for "smoke" the results are almost all within the past 2 years ...

I blame the hidden context on the tools/subagents. One recent example.. codex can just look in the code for Db schema but continually tries to request permission for a live db query