Claude Code Relaunches Projects with Multi-Agent Coordination
Anthropic has released a major update to Claude Code's Projects feature, introducing a multi-agent architecture that allows developers to manage several AI agents working simultaneously on the same codebase.
The new system centers on "threads," each of which runs as a separate Claude Code cloud session operating on its own branch and copy of the repository. A "coordinator" agent sits above these threads, organizing their work and maintaining shared memory, goals, and a common library of files and artifacts.
When multiple threads work on overlapping code, the system handles the conflict as a standard merge conflict—similar to what would occur in a pull request from two developers. Each thread can also further delegate tasks by splitting its assigned work into additional subtasks.
The approach mirrors capabilities offered by tools like Grok Bot, positioning Claude Code's Projects as a competitive option for teams looking to orchestrate coordinated AI-assisted development workflows in the cloud.