What's new in Claude

The stream-stall hint now reads "Waiting for API response · will retry in …" instead of "No response from API · Retrying in …", and triggers after 20 seconds of silence instead of 10.

Improved auto mode safety blocks destructive git and terraform/pulumi destroy commands; adds model deprecation warnings, fixes connection drops mid-stream to preserve partial responses, and resolves fullscreen TUI corruption and silent thinking-block failures.

New /config key=value inline syntax for settings, sandbox.allowAppleEvents for macOS, prompt caching fixes for custom API gateways, resolved Write/Edit truncation on network drives, and long paragraphs now stream line-by-line.

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

Sub-agents can spawn sub-agents — background agent chains now support up to five levels deep, enabling more complex autonomous workflows.

--safe-mode — start Claude Code with all customizations (hooks, plugins, MCPs) disabled for clean troubleshooting sessions.

fallbackModel — configure up to three fallback models tried in order when the primary model is unavailable or rate-limited.

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

Paid Claude plans now receive a dedicated monthly credit for programmatic usage, covering the Claude Agent SDK, claude -p CLI, Claude Code GitHub Actions, and third-party apps built on the Agent SDK.

The code execution tool now supports code_execution_20260521, disclosing the 90-second per-cell execution limit in the tool description so Claude can budget long-running cells.

Web search and web fetch tools add a response_inclusion parameter to drop consumed result blocks from API responses in agentic workflows — no beta header required.

Claude Code weekly limits are increasing 50%, now through July 13. Live now for all Pro, Max, Team, and seat-based Enterprise users. Compute headroom from the SpaceX partnership enabled the bump.

Doubling Claude Code's 5-hour rate limits for Pro, Max, and Team plans and removing the peak-hours limit reduction — effective immediately. A new SpaceX compute partnership is what made this possible.

New in Claude Code: Code Review. A team of agents runs a deep review on every PR. Code output per Anthropic engineer is up 200% this year and reviews were the bottleneck — personally been using it for weeks and it catches real bugs.

Claude Code can now upload and edit HTML artifacts that you can share with your team or other Claudes — starting with Team plans, coming to Pro and Max soon. Also: dynamic workflows let Claude write its own orchestration harness on the fly.