AI-generated editorial illustration; not a documentary photograph.
Anthropic has relaunched Projects in Claude Code as a way to coordinate multiple coding agents within one workspace. The beta became available on September 17, 2026, to select Claude Pro and Max subscribers, according to The Verge.
Parallel Work Through Threads
A project can contain multiple threads that handle different tasks concurrently. These threads share the project's memory, goals, files and artifacts, while a coordinator organizes their work and provides a central place for users to monitor progress or issue updates.
Users can also communicate with an individual thread when a task needs direct attention. Threads may divide delegated work further through subagents, loops and workflows, allowing larger assignments to be split into smaller pieces.
Separate Branches, Familiar Conflicts
Each thread operates as an independent Claude Code cloud session with its own repository copy and branch. That separation lets agents work in parallel, but it does not eliminate integration problems: when threads modify overlapping code, developers must resolve the resulting merge conflict as they would with competing pull requests.
Cloud First
The beta initially runs its threads in the cloud. Anthropic says support for local tools and code is due soon, but the source does not provide a specific release date. Until then, teams whose workflows depend on local environments will not have that capability within the revamped Projects system.
Access Remains Limited
Initial availability is restricted to selected Claude Pro and Max subscribers. Anthropic plans to expand Projects to all Pro, Max, Team and Enterprise users later, as well as to Cowork and standard Claude chats, though no schedule for that wider rollout was reported.
Source: Claude Code relaunches Projects to manage multiple AI agents in the cloud
SIGNAL TITANS
Leave a Reply