Anthropic updates Claude Code with cloud-based multi-agent projects

Anthropic has introduced a revamped projects feature for Claude Code, its developer-focused assistant. The update lets users manage a team of artificial intelligence agents that coordinate with one another to tackle complex software development tasks.

How multi-agent projects work

Under the new system, each project acts as a shared workspace equipped with a common memory, a set of goals, and a library of files and artifacts. A central coordinator directs the overall effort while individual threads run separate tasks in parallel.

Every thread operates within its own cloud session, working on a dedicated branch and copy of the code repository. If multiple threads modify the same code, the system resolves the overlap as a standard merge conflict. Furthermore, individual threads can split assigned work into smaller pieces using subagents, loops, and workflows to speed up completion.

Availability and rollout

Users can interact with each thread individually or oversee the entire operation through a main project chat. At launch, these threads run entirely in the cloud, though local tool and code support is scheduled for release soon.

The updated projects feature is currently available in beta for select Claude Pro and Max subscribers. Anthropic plans to expand access to all Pro, Max, Team, and Enterprise users, as well as Cowork and regular Claude chats, in the near future.