AI slop

[SlopBarking]

It consumes significantly more tokens than the vanilla version while delivering worse results.
Please don't use AI slop.

[code-yeongyu]

Fair feedback, and thanks for taking the time to say it. Token cost vs. value is a real tradeoff and not the right one for everyone.

Concretely: the heavier subagent/team fan-out is opt-in (plain sessions don't spawn subagents unless a mode like ultrawork/teammode/start-work orchestrates them), the Fast tier and CodeGraph daemon are opt-in/opt-out, and if omo isn't for you, npx lazycodex-ai uninstall removes it cleanly. We're actively tracking token-efficiency as a priority. Closing as answered - reopen with a specific scenario (which command, how many tokens, expected vs actual) if you want us to dig into a concrete regression.