Anthropic relaunched its projects feature in Claude Code. The update lets users run multiple AI agents together under one roof. These agents share memory, goals, and a library of files and artifacts.
This tool functions similarly to Grok Bot and other platforms that manage groups of artificial intelligence agents. Each project utilizes threads to run different tasks in parallel. A coordinator directs the entire operation.
Under the hood, each thread is a Claude Code cloud session. It works on its own branch and copy of the repository. The coordinator keeps all work organized.
If multiple threads work on the same code, the system resolves the overlap as a merge conflict. This process mirrors a standard pull request. Each thread can further split delegated work into smaller pieces. It uses subagents, loops, and workflows to finish large assignments faster.
Users can interact with each thread individually. They can also monitor and update activities through the main project chat.
At launch, these threads run exclusively in the cloud. However, Anthropic states that support for local tools and code is coming very soon.
The updated projects feature is available in beta starting today. Access is initially limited to select Claude Pro and Max subscribers.
Anthropic plans to expand access later. The rollout will eventually cover all Pro, Max, Team, and Enterprise users. It will also reach Cowork and regular Claude chats.
