Anthropic updates Projects in Claude Code to coordinate AI agents — The Verge
Anthropic has updated the Projects feature in Claude Code: it enables users to manage multiple AI agents that work on tasks in the cloud and coordinate with one another. As The Verge reports, agents within a project have shared memory, goals, and a library of files and artifacts.
Parallel work on tasks
Separate “streams” are created in each project and can perform different tasks in parallel. Their work is directed by a coordinator. Each stream is effectively a cloud Claude Code session that operates with its own branch and copy of the repository.
If several streams modify the same code, overlaps will be resolved as merge conflicts, just as when working with a pull request. Streams can also divide assigned tasks into smaller parts using subagents, loops, and workflows.
More current news is available on the UA.News Telegram channel Telegram.
Beta access
Users can interact with each stream separately or monitor their work and make changes through the project’s main chat. At launch, streams operate in the cloud, while Anthropic plans to add support for local tools and code in the near future.
The updated feature is available in beta for select Claude Pro and Max subscribers. Access is later expected to expand to all users of Pro, Max, Team, and Enterprise plans, as well as to Cowork and regular Claude chats.
