Claude Code Gets Major Update: AI Sessions Can Now Communicate Autonomously, Marking a Key Step in Multi-Agent Collaboration

Anthropic announced on Friday that its AI coding tool Claude Code has received a major update allowing AI agents across different sessions to communicate autonomously. The company said the cross-session messaging feature enables agents running in multiple terminal windows to align on progress, share discoveries and automatically warn about code conflicts without manual context copying. The feature requires Claude Code v2.1.224 or higher and runs on macOS and Linux systems. When a session meets version requirements the feature is enabled by default. Anthropic said the update addresses the biggest pain point in multi-task parallel development by eliminating manual context synchronization between sessions handling frontend, backend or different Git Worktree branches. Anthropic also announced that starting August 14 for Pro, Max and Team subscribers, Claude Code's default permission mode will shift to Auto Mode. The company said Auto Mode uses an independent AI classifier to evaluate every tool call and shell command in real time, allowing safe operations while blocking destructive or unauthorized actions. In tests with 1,053 paying beta testers, traditional manual approval identified 13.6% of dangerous commands while Auto Mode identified 89%.