This agent teams feature is a big step forward vibecoding claude opus

Anthropic released agent teams inside of Cloud Code yesterday, and here is why this is so beneficial. Before, Cloud Code would just spawn sub-agents that receive a prompt from the parent agent and then send a response back, but this allows you to send out a prompt to the agent and then it splits out into different terminals, each with their own task, and then they can communicate like teammates under the hood, with everything being controlled in one team-up session. In the documentation that Anthropic released, they say that sub-agents are best for focused tasks where only the results matter, and agent teams is best for complex work requiring discussion and collaboration. If you guys are wanting to give this a try, they do say that agent teams is disabled by default, but in order to enable them, you will have to set your Cloud Code experimental agent teams environment variable to 1, either in your shell environment or through settings.json. Honestly, the results that I've seen from this are insane. Just the way that it splits it up into a full dev team and you can have full control of each agent, it really just feels like a step forward in this vibe coding journey. Give me a follow and join the discord in the link in my bio if you want to keep up with this AI space.