Claude Code Projects now runs parallel AI workers with less babysitting

Anthropic has updated Claude Code Projects to make managing parallel cloud sessions less hands-on. According to the report, the changes include proactive coordinator messages, read-only thread sharing, reminders, branching commands and improved artifact navigation. The goal is to cut down the screen time needed to keep track of several sessions running at once.

Here is how it works: each project is a long-running conversation with a coordinator that assigns work to threads. Every thread runs as an independent Claude Code cloud session with its own repository copy and Git branch. The coordinator steers the project from thread reports, while overlapping code changes surface as ordinary Git merge conflicts when branches are combined. According to the article, the coordinator now answers questions directly in project chat and posts status updates as threads progress, so pending decisions, completed tasks and blocked work appear in the main conversation rather than in a thread list.

Sharing comes with a limit: threads can be shared through read-only links with anyone signed in to Claude, without exposing the rest of the project. Threads that accessed a local computer stay unshareable. The report says the restriction keeps shared transcripts from exposing local file contents, shell output or other machine data gathered during the session.

Projects carry no separate fee, but every worker draws on the subscriber's existing Claude Code allowance, and accounts are capped at 200 new threads per day across all projects. New projects default to Opus for both roles, with low effort for the coordinator and high effort for workers. Running several high-effort sessions at once can use up plan limits faster than a single Claude Code session.