THE CRUNCH
Anthropic has relaunched the Projects feature in Claude Code, allowing users to manage multiple AI agents in the cloud. The revamped tool lets users run several agents under one roof, sharing memory, goals, and a library of files and artifacts. Each project has threads running different tasks in parallel, with a coordinator directing the work. Under the hood, each thread is a separate Claude Code cloud session on its
own branch and copy of the repo. The coordinator keeps work organised, but if threads work on the same code, overlap is resolved as a merge conflict, just like a pull request. This setup mirrors tools such as Grok Bot that manage groups of AI agents.
The feature aims to simplify complex workflows by keeping all agents and their outputs in one place. By treating each thread like a separate session with its own branch, Anthropic lets developers track changes and resolve conflicts without leaving the Projects interface.