Claude Code relaunches Projects to manage multiple AI agents in the cloud

Claude Code’s revamped Projects feature lets users run multiple AI agents within one workspace, sharing memory, goals, files, and artifacts. Each project can split work into parallel threads, with a coordinator managing tasks while individual cloud sessions operate on separate repo branches.

The feature reflects a broader shift from single-agent coding assistants toward multi-agent orchestration for software development. By treating AI workstreams more like parallel pull requests, Anthropic is aligning agentic coding tools with existing developer workflows while exposing familiar challenges like merge conflicts.