the latest updates to your ai tools, straight from their official changelogs.
~Every API response now includes the exact cost of the request via a cost_in_usd_ticks field in the usage object.
Every API response now includes the exact cost of the request via a cost_in_usd_ticks field in the usage object. Works across chat completions, Responses API, image generation, video generation, and streaming. For more details, see the Cost Tracking docs.
continue reading on the release page →~Cursor Security Review is now in beta on Teams and Enterprise plans.
Cursor Security Review is now in beta on Teams and Enterprise plans. You can run two types of always-on security agents: Security Reviewer and Vulnerability Scanner.
Security Reviewer checks every PR for security vulnerabilities, auth regressions, privacy and data-handling risks, agent tool auto-approvals, and prompt injection attacks. It will leave inline comments at the exact diff location with severity and remediation.
Vulnerability Scanner runs scheduled scans of your codebase to check for known vulnerabilities, outdated dependencies, and configuration issues. You can configure it to send updates of its findings in Slack.
Customize Cursor-managed security agents by adjusting triggers, adding your own instructions, giving them custom tooling, and choosing how outputs are shared. For example, you can plug in MCP servers for your existing SAST, SCA, and secrets scanners for Cursor to use as part of the review.
We're also continuously improving the runtime, harness, and models powering Cursor Security Review for a strong out-of-the-box experience.
Security agents draw from your existing usage pool. Admins can enable Security Review in the Cursor dashboard to get started.
continue reading on the release page →
ANTHROPIC_BEDROCK_SERVICE_TIER environment variable to select a Bedrock service tier (`default`, `flex`, or `priority`), sent as the `X-Amzn-Bedrock-Service-Tier` header/mcp now shows claude.ai connectors hidden by a manually-added server with the same URL, with a hint to remove the duplicate/mcp message shown when an MCP server is still unauthorized after the browser sign-in flowclaude_code.at_mention log event for `@`-mention resolution/branch producing forks that fail with "tool_use ids were found without tool_result blocks" when the source session contained entries from rewound timelines/model not showing the Effort option for Bedrock application inference profile ARNs, and those ARNs not receiving `output_config.effort`spinnerTipsOverride.excludeDefault not suppressing the time-based spinner tips!exit / `!quit` in bash mode terminating the CLI instead of running as a shell command+You can now set an expiration policy on uploaded files using expires_after or an explicit expires_at timestamp.
You can now set an expiration policy on uploaded files using expires_after or an explicit expires_at timestamp. Expired files are automatically deleted. For more details, see the Files API docs.
continue reading on the release page →alwaysLoad option to MCP server config — when `true`, all tools from that server skip tool-search deferral and are always availableclaude plugin prune to remove orphaned auto-installed plugin dependencies; `plugin uninstall --prune` cascadesclaude -p: `CLAUDE_CODE_FORK_SUBAGENT=1` now works in non-interactive sessions--dangerously-skip-permissions no longer prompts for writes to `.claude/skills/`, `.claude/agents/`, and `.claude/commands/`/terminal-setup now enables iTerm2's "Applications in terminal may access clipboard" setting so `/copy` works, including from tmux/usage leaking up to ~2GB of memory on machines with large transcript histories--resume crashing on startup in external builds--resume failing on large sessions when a transcript line was corrupted by an unclean shutdown — the corrupt line is now skippedthinking.type.enabled is not supported error when using Bedrock application inference profile ARNsclaude ultrareview [target] subcommand to run `/ultrareview` non-interactively from CI or scripts — prints findings to stdout (`--json` for raw output) and exits 0 on completion or 1 on failureAI_AGENT environment variable for subprocesses so `gh` can attribute traffic to Claude Codeclaude plugin validate now accepts `$schema`, `version`, and `description` at the top level of `marketplace.json` and `$schema` in `plugin.json`/usage now opens the native Account & Usage dialog instead of returning plain-text session cost/rewind and other interactive overlays not responding to keyboard input after launching with `claude --resume`DISABLE_TELEMETRY / `CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC` not suppressing usage metrics telemetry for API and enterprise users+This release introduces a new way to multitask with async subagents, an improved worktrees experience, and multi-root workspaces for making cross-repo changes.
This release introduces a new way to multitask with async subagents, an improved worktrees experience, and multi-root workspaces for making cross-repo changes.
With /multitask, Cursor will run async subagents to parallelize your requests instead of adding them to the queue. It will also break down larger tasks into smaller chunks for a fleet of async subagents to tackle simultaneously.
If you already have messages in the queue, you can ask Cursor to multitask on them instead of waiting for the current run to finish.
We've added new and improved worktrees in the Agents Window.
Run isolated tasks in the background across different branches. When you're ready to test changes, move any branch into your local foreground with one click.
A single agent session can now target a reusable workspace made of multiple folders.
continue reading on the release page →prUrlTemplate setting to point the footer PR badge at a custom code-review URL instead of github.comCLAUDE_CODE_HIDE_CWD environment variable to hide the working directory in the startup logo/config settings (theme, editor mode, verbose, etc.) now persist to `~/.claude/settings.json` and participate in project/local/policy override precedence--from-pr now accepts GitLab merge-request, Bitbucket pull-request, and GitHub Enterprise PR URLs--print mode now honors the agent's `tools:` and `disallowedTools:` frontmatter, matching interactive-mode behavior--agent <name> now honors the agent definition's `permissionMode` for built-in agentsPostToolUse and `PostToolUseFailure` hook inputs now include `duration_ms` (tool execution time, excluding permission prompts and PreToolUse hooks)blockedMarketplaces now correctly enforces `hostPattern` and `pathPattern` entriesPostToolUse hooks that emit no response payload writing empty entries to the session transcript+You can now use grok-voice-think-fast-1.0 with the Speech to Speech API.
You can now use grok-voice-think-fast-1.0 with the Speech to Speech API. To get started, check out the Speech to Speech docs. For more details, see our blog post.
continue reading on the release page →v) and visual-line mode (`V`) with selection, operators, and visual feedbackDISABLE_UPDATES env var to completely block all update paths including manual `claude update` — stricter than `DISABLE_AUTOUPDATER`claude plugin tag to create release git tags for plugins with version validation/cost and `/stats` into `/usage` — both remain as typing shortcuts that open the relevant tab"$defaults" in `autoMode.allow`, `autoMode.soft_deny`, or `autoMode.environment` to add custom rules alongside the built-in list instead of replacing it--continue/`--resume` now find sessions that added the current directory via `/add-dir`/color now syncs the session accent color to claude.ai/code when Remote Control is connected/model picker now honors `ANTHROPIC_DEFAULT_*_MODEL_NAME`/`_DESCRIPTION` overrides when using a custom `ANTHROPIC_BASE_URL` gateway/mcp menu hiding OAuth Authenticate/Re-authenticate actions for servers configured with `headersHelper`, and HTTP/SSE MCP servers with custom headers being stuck in "needs authentication" after a transient 401mcpServers are now loaded for main-thread agent sessions via `--agent`/resume command now offers to summarize stale, large sessions before re-reading them, matching the existing `--resume` behaviorplugin install on an already-installed plugin now installs any missing dependencies instead of stopping at "already installed"blockedMarketplaces and `strictKnownMarketplaces` are now enforced on plugin install, update, refresh, and autoupdatecleanupPeriodDays retention sweep now also covers `~/.claude/tasks/`, `~/.claude/shell-snapshots/`, and `~/.claude/backups/`user_prompt events now include `command_name` and `command_source` for slash commands; `cost.usage`, `token.usage`, `api_request`, and `api_error` now include an `effort` attribute when the model supports effort levels. Custom/MCP command names are redacted unless `OTEL_LOG_TOOL_DETAILS=1` is setwhere.exe executable lookups per process for faster subprocess launches/model: selections now persist across restarts even when the project pins a different model, and the startup header shows when the active model comes from a project or managed-settings pinWebFetch hanging on very large HTML pages by truncating input before HTML-to-markdown conversion/login having no effect when launched with `CLAUDE_CODE_OAUTH_TOKEN` env var and that token expiresCtrl+_) doing nothing immediately after typing, and skipping a state on each undo stepNO_PROXY not being respected for remote API requests when running under Bunreload_plugins reconnecting all user MCP servers serially/resume on large sessions is significantly faster (up to 67% on 40MB+ sessions) and handles sessions with many dead-fork entries more efficiently/config search now matches option values (e.g. searching "vim" finds the Editor mode setting)/doctor can now be opened while Claude is responding, without waiting for the current turn to finish/reload-plugins and background plugin auto-update now auto-install missing plugin dependencies from marketplaces you've already addedhooks: now fire when running as a main-thread agent via `--agent`sandbox.network.deniedDomains setting to block specific domains even when a broader `allowedDomains` wildcard would otherwise permit themCtrl+A and `Ctrl+E` now move to the start/end of the current logical line in multiline input, matching readline behaviorCtrl+Backspace now deletes the previous word/extra-usage now works from Remote Control (mobile/web) clientscd <current-directory> && git … no longer triggers a permission prompt when the `cd` is a no-op/private/{etc,var,tmp,home} paths are now treated as dangerous removal targets under `Bash(rm:*)` allow rulesBash(find:*) allow rules no longer auto-approve `find -exec`/`-delete`/loop: pressing Esc now cancels pending wakeups, and wakeups display as "Claude resuming /loop wakeup" for clarity/ultrareview: faster launch with parallelized checks, diffstat in the launch dialog, and animated launching stateCtrl+U to once again delete from the cursor to the start of the line/copy "Full response" not aligning markdown table columns for pasting into GitHub, Notion, or SlackdangerouslyDisableSandbox running commands outside the sandbox without a permission prompt/effort auto confirmation — now says "Effort level set to max" to match the status bar labelxhigh effort level for Opus 4.7, sitting between `high` and `max`. Available via `/effort`, `--effort`, and the model picker; other models fall back to `high`/less-permission-prompts skill — scans transcripts for common read-only Bash and MCP tool calls and proposes a prioritized allowlist for `.claude/settings.json`/ultrareview for running comprehensive code review in the cloud using parallel multi-agent analysis and critique — invoke with no arguments to review your current branch, or `/ultrareview <PR#>` to fetch and review a specific GitHub PROTEL_LOG_RAW_API_BODIES environment variable to emit full API request and response bodies as OpenTelemetry log events for debugging/effort now opens an interactive slider when called without arguments, with arrow-key navigation between levels and Enter to confirm/setup-vertex and `/setup-bedrock` to show the actual `settings.json` path when `CLAUDE_CONFIG_DIR` is set, seed model candidates from existing pins on re-run, and offer a "with 1M context" option for supported models@ file suggestions re-scanning the entire project on every turn in non-git working directories, and showing only config files in freshly-initialized git repos with no tracked files/resume immediately resuming an arbitrary titled session instead of showing the session picker/context grid rendering with extra blank lines between rows/clear dropping the session name set by `/rename`, causing statusline output to lose `session_name`/tui command and `tui` setting — run `/tui fullscreen` to switch to flicker-free rendering in the same conversationautoScrollEnabled config to disable conversation auto-scroll in fullscreen modeCtrl+O to toggle between normal and verbose transcript only; focus view is now toggled separately with the new `/focus` command--resume/`--continue` now resurrects unexpired scheduled tasks/context, `/exit`, and `/reload-plugins` now work from Remote Control (mobile/web) clientsdisable-model-invocation: true failing when invoked via `/<skill>` mid-message--resume sometimes showing the first prompt instead of the `/rename` name for sessions still running or exited uncleanly