What's new in Claude

Fixed mid-stream connection drops, mouse-wheel scrolling in WSL2, sandbox glob performance issues, feedback survey capture, welcome screen stacking, and various UI/remote session improvements.

Added Tool(param:value) syntax for permission rules, improved nested .claude/ directory handling, enhanced auto mode subagent evaluation, and fixed multiple subagent transcript and MCP server issues.

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

Sub-agent chains — Sub-agents can now spawn their own sub-agents; background chains are capped at five levels deep to prevent runaway recursion.

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

fallbackModel — Configure up to three fallback models tried in sequence if the primary model is unavailable or over capacity.

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

Claude Platform on AWS: The GET /v1/environments/{id}/work endpoint for self-hosted sandboxes is now available on Claude Platform on AWS.

Claude Fable 5 launched — The most capable widely released model: 1M token context window, 128k max output tokens, always-on adaptive thinking, and a new stop_reason: "refusal" with zero billing on pre-output refusals. Claude Mythos 5 available for Project Glasswing participants.

Claude Code weekly limits raised 50% through July 13, 2026 for Pro, Max, Team, and Enterprise plans. Starting June 15, paid plan users can claim a dedicated monthly credit for programmatic usage covering the Agent SDK, claude -p CLI, GitHub Actions, and third-party Agent SDK apps.

Claude Fable 5 is here — Anthropic's most capable widely released model, sitting above the Opus class with a 1M token context window and always-on adaptive thinking. Claude Mythos 5 also launched for Project Glasswing participants.

Boris Cherny (Claude Code creator) says he hasn't written a line of code himself in 2026. Shared tips from his surprisingly vanilla setup: he runs 5 parallel Claude Code instances using separate git checkouts of the same repo, and focuses on teaching accessible AI agent usage.

Thariq called Workflows "the biggest upgrade to Claude Code's capabilities since skills and subagents," highlighting how they enable non-technical task orchestration — diving deep with the team on best practices and example patterns.