It seemed that with the emergence of AI agents, a developer's life would become easier - you set a task and relax. In reality, it's the opposite. The speed has increased, the possibilities have grown, and now everyone is simultaneously handling 6+ features and projects. It's overwhelming to keep track of what's currently in progress, and there's no convenient tool to manage all of this.
Ideally, several agents should work on one feature at the same time - one plans the architecture, another writes code, a third critiques decisions, and a fourth tests. And most likely, these will be agents from different manufacturers, each with their own strengths.
IDEs have long ceased to be just text editors - they have become a code management environment. The next step is obvious: from a code management environment to an agent management environment. Whoever first creates a convenient “dispatcher” for orchestrating dozens of agents will take the market.