What's new in Claude

Added /rewind support, fixed scroll position jumping and background agent resurrection, improved MCP reliability, and reduced CPU usage by ~37%.

Bug fixes and reliability improvements.

Added sandbox.credentials setting, org-configured model restrictions, mouse click support in fullscreen, and fixed structured output and remote MCP tool call timeout issues.

Added claude mcp login/logout CLI commands, status filtering to workflows, skills section to /plugin, and ! bash commands now trigger auto-response.

/cd Command — Move the current session to a new working directory mid-conversation without rebuilding the prompt cache.

Sub-agents spawning sub-agents — Background chains can now go up to five levels deep, up from the previous limit.

--safe-mode flag — Start Claude Code with all customizations disabled for clean troubleshooting.

fallbackModel setting — Configure up to three fallback models tried in order when the primary model is unavailable.

Version requirements for managed deployments — Admins can now require an approved Claude Code version range across the organization.

Model retirements — Claude Sonnet 4 (claude-sonnet-4-20250514) and Claude Opus 4 (claude-opus-4-20250514) have been retired — all requests will now return an error. Upgrade to Claude Sonnet 4.6 and Claude Opus 4.8 respectively.

Claude Tag launch: "Claude Tag is Claude Code made multiplayer, async, and proactive across your whole team. In beta today for Claude Enterprise and Team plans." The team noted it now writes 65% of Anthropic's product team's code.

"Claude Tag is an evolution of Claude Code, made more proactive and built to work with a full team. It's now one of the main ways we get things done at Anthropic: 65% of our product team's code now comes from our internal version."

"Cool way to use Claude Code: deciphering Linear A, a 3500 year old written language from Crete 🤞" — Boris Cherny, Claude Code creator.