I tried the “New Agent” button, and asked Claude it, and got this reply:-
## About spawning additional agents
No, I don't have a tool to spawn additional agents.
What’s the point of this new idea? Other than putting my “tabs” in a vertical list, it seems like “chat” still?
Is this planned to do something special one day?
The “blurb” (https://cursor.com/blog/2-0) says this:-
Cursor 2.0 makes it easy to run many agents in parallel without them interfering with one another, powered by git worktrees or remote machines. We’ve even found that having multiple models attempt the same problem and picking the best result significantly improves the final output, especially for harder tasks.
which is still nothing new over what we could do before (you could have as many open chat tabs doing stuff in parallel as you like).
I was under the impression that the agent was an orchestrator - but it has no way to do this - e.g. “multiple models attempt the same problem and picking the best result” is this marketing feature creep where we the humans are the ones who have to manually do all that model picking and result testing ourselves?
What am I missing?