There's some real research-backed evidence that this approach elevates output quality, and it was even discussed a few months back: https://news.ycombinator.com/item?id=47169627
> We expect that agents coordinating in the wild will act in higher variance ways than we see here, because they’ll have different backgrounds and therefore different contexts. They also, presumably, won’t all be Claudes.
Claude chat itself called it 'Claude-on-Claude action', which I found cute.