반복만으로는 루프가 아닙니다 —
'아니오'라고 말해 줄 신호가 있어야 나아지고, 그 신호는 만든 쪽이 아니라 다른 쪽이 줘야 합니다.
Repetition alone isn't a loop — it improves only with a signal that can say 'no,' and that signal must come from someone other than the maker.
ai-toolingcontext-engineering
반복만으로는 루프가 아닙니다 —
'아니오'라고 말해 줄 신호가 있어야 나아지고, 그 신호는 만든 쪽이 아니라 다른 쪽이 줘야 합니다.
루프는 목표·실행·신호·반복 네 부품으로 됩니다. 그중 신호 — 검증(Evaluation) — 이 빠지면, 백 번을 돌려도 결과가 나아지지 않습니다. 그래서 /goal은 매 턴 끝마다 별도의 작은 모델이 끝났는지 판정하게 합니다. 자기 숙제를 자기가 채점하지 않는, 만드는 쪽과 채점하는 쪽의 분리가 루프의 심장입니다.Also in EnglishRepetition alone isn't a loop — it improves only with a signal that can say 'no,' and that signal must come from someone other than the maker.
A loop has four parts: goal, execution, signal, repetition. Drop the signal — the evaluation — and a hundred turns won't improve the result. That's why /goal has a separate small model judge whether each turn is done. Separating the maker from the grader, so the agent never grades its own homework, is the heart of the loop.