agents/critic.py

from app import llm_model

def review_result(step, result):
    prompt = f"Review the result of this step:\nStep: {step}\nResult: {result}\nIs it correct?"
    return llm_model(prompt)