At the moment - it's model's judgement upon search results for every claim from provided content. If search results returned inaccurate data, or were fabricated

At the moment - it's model's judgement upon search results for every claim from provided content. If search results returned inaccurate data, or were fabricated - it might affect a verdict for that claim, or highlight that it's unverifiable or contradictory. But there is a room for improvement, what do you suggest? Have a Judge agent which will check results?