Big stability & performance pass: faster startup, binary shrinks 340MB→75MB, lower memory, MCP auto-reconnects, plus 33 fixes for hangs, auto mode, VS Code Focus view, and Chrome.
/usage: per-loop run count, total tokens, tokens per run, and last run, so runaway or chatty /loop tasks are easy to spotmodelPicker setting: curate the /model picker with an ordered, labeled list of models (any id spelling, including Vertex/Bedrock ids), appended to or replacing the built-in lineuppromptCacheTtl and subagentPromptCacheTtl settings so API-key and cloud-provider users can keep a 1-hour prompt cache on the main conversation while subagents stay at 5 minutesmodelPricing managed setting so an organization's contracted per-model rates and discount multiplier are used for /cost, the status line, and telemetry cost figures instead of list price/login → Anthropic Console: "Sign in with your Console account" (recommended) alongside creating an API key, so organizations that don't allow API keys can sign inSkipped sources line to /status that lists managed settings sources (for example managed-settings.json) present but not applied because a higher-precedence managed source is activemanaged marker in /mcp and /plugins on claude.ai connectors whose authentication is managed by your organization/web-setup/status line showing whether GitHub is connected for Claude Code on the web (Pro/Max), pointing to /web-setup when it isn't/tasks and the agent detail dialogsANTHROPIC_AUTH_TOKEN directly against the Anthropic API, so its data-handling settings apply/login over SSH: the sign-in URL appears immediately, pressing c reports how the URL was copied instead of always claiming success, and a hint explains how to select text in fullscreenxhigh/max is used with thinking turned off: it now names the level, the setting that disabled thinking, and /effort high as the fix/loop: consecutive wake-ups where Claude has nothing to do now fold into a single line in the terminal instead of printing each one-p) and SDK sessions never recovering after a dropped connection; they now reconnect automatically or report as failed/model picker silently ignoring an Ultracode selection; picking Ultracode now applies it to the current session/resume only listing the 50 most recent sessions; the picker now loads more as you scroll/login) losing its leading columns when another part of the screen repaintsspellcheck not underlining a misspelled word typed directly after an emojiAPI Error: No response from APIcompanyAnnouncements not showing at startup in a session that began with signing in (for example the first launch after /logout)if conditions like Bash(cat *) firing on unrelated Bash commands when the command contained $() or backtick command substitution followed by more argumentsmarketplace field never resolving when both plugins are loaded together via --plugin-dir/reload-plugins keeping the LSP tool after the last LSP plugin is disabled; it now also warns before an LSP plugin change that would re-read the conversation--agents silently ignoring invalid JSON or invalid agent definitions; it now exits with a clear error, like --mcp-config/status showing "Found invalid entries in: ." with no filename when ~/.claude.json has an invalid MCP server entry/clear removing the /rename session name from the prompt bar even though the name was kept for the new session~/.claude/history.jsonl contains a malformed entryHTTPS_PROXY (and sometimes failing) when localhost was listed in NO_PROXY but not lowercase no_proxy; both casings are now honoredcurl printing the proxy's 403 page)rate_limits fields and /usage still showing a rate-limit window's pre-reset usage percentage after the window reset while the session was idleclaude --teleport <session> exiting on uncommitted changes instead of offering to stash them and continue, as the session picker already does/web-setup repeatedly asking you to log in when an older GitHub CLI (without gh auth token) was already authenticatedclaude launcherclaude remote-control exiting and stranding attached Remote Control sessions when the server drops its environment mid-session; it now recoversclaude remote-control sometimes getting stuck after it was stopped and restarted, for Team and Enterprise members without an admin or owner role/model picker and the bundled claude-api skill to show Sonnet 5's $2/$10 per Mtok pricing as its standard list price rather than a limited-time promo/model, /fast, and /effort to also run immediately instead of queueing until the turn ends on Bedrock, Vertex, and Foundry and when telemetry is disabled