Sub-agents can now spawn sub-agents 5 levels deep; major stability pass covers agent lifecycle, model picker accuracy, wildcard permissions, and memory recall in remote sessions.
/pluginmodel attribute to the claude_code.lines_of_code.count OTEL metric/goal status chip no longer re-renders the terminal at 5 Hz while idle, and fewer UI re-renders while subagents run in parallel.mcp.json approvals, trust) when dispatched onto a pre-warmed worker/model suggestions in the claude agents dispatch input rendering with a misleading slash prefix and showing models disabled for your orgavailableModels restrictions not being applied to subagent model overrides, the agent dispatch model picker, and the advisor modelavailableModels allowlists hiding the /model picker's Opus and Sonnet 1M rows when entries use version-specific IDs like claude-opus-4-8/model picker on Bedrock offering models the provider doesn't serve — selecting one silently switched the session model and lit the selection marker on multiple rows[1M][1m]) when ANTHROPIC_DEFAULT_OPUS_MODEL already includes oneopusplan model setting not shipping with 1M context in plan mode for entitled users; the opusplan[1m] workaround now also correctly switches to Opus in plan modeWebFetch(domain:*.example.com) wildcard domain rules never matching subdomains in allow, deny, and ask position, and file permission rules with mid-pattern wildcards (e.g. Read(secrets-*/config.json)) being rejected at startupCLAUDE_MEMORY_STORES) in remote sessionsDate.now()/Math.random()/plugin marketplace list losing its cursor after backing out of a long plugin list, and Esc from the plugin browser returning to the wrong tab~/.aws config files when AWS_REGION isn't set, matching AWS SDK precedence; /status shows where the region came from/code-review now keeps the ultra option visible when you're not signed in to claude.ai, with an explanation that the cloud review requires a claude.ai account/loop in remote sessions, where pending loops don't keep the container alive