Anthropic Relaunches Claude Code Projects to Orchestrate Multiple AI Agents

It is a relaunched feature that runs multiple AI agents under one project with shared memory, goals and a library of files and artifacts, coordinated by a 'coordinator' that delegates work across parallel threads.

A single session handles one task in one context, while Projects runs several parallel cloud sessions, each on its own branch and repo copy, with a coordinator unifying results and resolving conflicts.

They are resolved as a normal merge conflict, just like any pull request, meaning the team reviews and merges according to standard Git workflow.

It is worth piloting on large multi-task projects, but start in non-critical repos and enforce strict human review before merging, especially in government and regulated environments.