what changed.

the latest updates to your ai tools, straight from their official changelogs.

2026-02-20

v2.1.50claude

6 new 2 changed 3 improved 14 fixed

+Added support for startupTimeout configuration for LSP servers
+Added WorktreeCreate and `WorktreeRemove` hook events, enabling custom VCS setup and teardown when agent worktree isolation creates or removes worktrees.
+Added support for isolation: worktree in agent definitions, allowing agents to declaratively run in isolated git worktrees.

original ↗

+ 22 more− less

full release →

+Added claude agents CLI command to list all configured agents
+Added CLAUDE_CODE_DISABLE_1M_CONTEXT environment variable to disable 1M context window support
+VSCodeAdded /extra-usage command support in VS Code sessions

~CLAUDE_CODE_SIMPLE mode now also disables MCP tools, attachments, hooks, and CLAUDE.md file loading for a fully minimal experience.
~Opus 4.6 (fast mode) now includes the full 1M context window

↑Improved memory usage during long sessions by clearing internal caches after compaction
↑Improved memory usage during long sessions by clearing large tool results after they have been processed
↑Improved startup performance for headless mode (`-p` flag) by deferring Yoga WASM and UI component imports

✓Fixed a bug where resumed sessions could be invisible when the working directory involved symlinks, because the session storage path was resolved at different times during startup. Also fixed session data loss on SSH disconnect by flushing session data before hooks and analytics in the graceful shutdown sequence.
✓LinuxFixed native modules not loading on systems with glibc older than 2.30 (e.g., RHEL 8)
✓Fixed memory leak in agent teams where completed teammate tasks were never garbage collected from session state
✓Fixed CLAUDE_CODE_SIMPLE to fully strip down skills, session memory, custom agents, and CLAUDE.md token counting
✓Fixed /mcp reconnect freezing the CLI when given a server name that doesn't exist
✓Fixed memory leak where completed task state objects were never removed from AppState
✓Fixed bug where MCP tools were not discovered when tool search is enabled and a prompt is passed in as a launch argument
✓Fixed a memory leak where LSP diagnostic data was never cleaned up after delivery, causing unbounded memory growth in long sessions

6 more fixed on the release page →

feb 20feb 20

2026-02-19

v2.1.49claude

3 new 7 changed 7 improved 10 fixed

+Added --worktree (`-w`) flag to start Claude in an isolated git worktree
+Added Ctrl+F keybinding to kill background agents (two-press confirmation)
+Added ConfigChange hook event that fires when configuration files change during a session, enabling enterprise security auditing and optional blocking of settings changes.

original ↗

+ 24 more− less

full release →

~Subagents support isolation: "worktree" for working in a temporary git worktree
~Agent definitions support `background: true` to always run as a background task
~Plugins can ship settings.json for default configuration
~Simple mode (CLAUDE_CODE_SIMPLE) now includes the file edit tool in addition to the Bash tool, allowing direct file editing in simple mode.
~Permission suggestions are now populated when safety checks trigger an ask response, enabling SDK consumers to display permission options
~Sonnet 4.5 with 1M context is being removed from the Max plan in favor of our frontier Sonnet 4.6 model, which now has 1M context. Please switch in /model.
~SDK model info now includes `supportsEffort`, `supportedEffortLevels`, and `supportsAdaptiveThinking` fields so consumers can discover model capabilities.

↑Improved MCP OAuth authentication with step-up auth support and discovery caching, reducing redundant network requests during server connections
↑Improved performance in non-interactive mode (`-p`) by skipping unnecessary API calls during startup
↑Improved performance by caching authentication failures for HTTP and SSE MCP servers, avoiding repeated connection attempts to servers requiring auth
↑Improved startup performance by caching MCP auth failures to avoid redundant connection attempts
↑Improved startup performance by reducing HTTP calls for analytics token counting
↑Improved startup performance by batching MCP tool token counting into a single API call
↑Improved permission prompts for path safety and working directory blocks to show the reason for the restriction instead of a bare prompt with no context

✓Fixed file-not-found errors to suggest corrected paths when the model drops the repo folder
✓Fixed Ctrl+C and ESC being silently ignored when background agents are running and the main thread is idle. Pressing twice within 3 seconds now kills all background agents.
✓Fixed prompt suggestion cache regression that reduced cache hit rates.
✓Fixed plugin enable and `plugin disable` to auto-detect the correct scope when `--scope` is not specified, instead of always defaulting to user scope
✓Fixed verbose mode not updating thinking block display when toggled via `/config` — memo comparators now correctly detect verbose changes
✓Fixed unbounded WASM memory growth during long sessions by periodically resetting the tree-sitter parser
✓Fixed potential rendering issues caused by stale yoga layout references
✓Fixed unbounded memory growth during long-running sessions caused by Yoga WASM linear memory never shrinking

2 more fixed on the release page →

feb 19feb 19

2026-02-18

v2.1.47claude

3 new 6 changed 6 improved 53 fixed

+Added last_assistant_message field to Stop and SubagentStop hook inputs, providing the final assistant response text so hooks can access it without parsing transcript files.
+Added chat:newline keybinding action for configurable multi-line input (anthropics/claude-code#26075)
+Added added_dirs to the statusline JSON `workspace` section, exposing directories added via `/add-dir` to external scripts (anthropics/claude-code#26096)

original ↗

+ 65 more− less

full release →

~Search patterns in collapsed tool results are now displayed in quotes for clarity
~Use ctrl+f to kill all background agents instead of double-pressing ESC. Background agents now continue running when you press ESC to cancel the main thread, giving you more control over agent lifecycle.
~Simplified teammate navigation to use only Shift+Down (with wrapping) instead of both Shift+Up and Shift+Down.
~Increased initial session count in resume picker from 10 to 50 for faster session discovery (anthropics/claude-code#26123)
~Moved config backup files from home directory root to `~/.claude/backups/` to reduce home directory clutter (anthropics/claude-code#26130)
~The /rename command now updates the terminal tab title by default (anthropics/claude-code#25789)

↑Improved VS Code plan preview: auto-updates as Claude iterates, enables commenting only when the plan is ready for review, and keeps the preview open when rejecting so Claude can revise.
↑Improved memory usage in long-running sessions by releasing API stream buffers, agent context, and skill state after use
↑Improved startup performance by deferring SessionStart hook execution, reducing time-to-interactive by ~500ms.
↑Improved performance of @ file mentions - file suggestions now appear faster by pre-warming the index on startup and using session-based caching with background refresh.
↑Improved memory usage by trimming agent task message history after tasks complete
↑Improved memory usage during long agent sessions by eliminating O(n²) message accumulation in progress updates

✓Fixed FileWriteTool line counting to preserve intentional trailing blank lines instead of stripping them with `trimEnd()`.
✓Fixed Windows terminal rendering bugs caused by `os.EOL` (`\r\n`) in display code — line counts now show correct values instead of always showing 1 on Windows.
✓Fixed a bug where bold and colored text in markdown output could shift to the wrong characters on Windows due to `\r\n` line endings.
✓Fixed compaction failing when conversation contains many PDF documents by stripping document blocks alongside images before sending to the compaction API (anthropics/claude-code#26188)
✓Fixed an issue where bash tool output was silently discarded on Windows when using MSYS2 or Cygwin shells.
✓Fixed the bash permission classifier to validate that returned match descriptions correspond to actual input rules, preventing hallucinated descriptions from incorrectly granting permissions
✓Fixed user-defined agents only loading one file on NFS/FUSE filesystems that report zero inodes (anthropics/claude-code#26044)
✓Fixed plugin agent skills silently failing to load when referenced by bare name instead of fully-qualified plugin name (anthropics/claude-code#25834)

45 more fixed on the release page →

feb 18feb 18

CLI Improvements and Mermaid ASCII Diagrams

cursor

2 new 22 changed 1 improved 17 fixed

+Added Emacs-style navigation: Ctrl+N/Ctrl+P for up/down and Ctrl+G for cancel/dismiss, alongside the existing arrow keys and Esc.
+Added keybindings: Alt+Delete deletes the previous word; Ctrl+D exits questionnaire prompts.
~AI code and conversations in CLI are now tracked by Cursor Blame.

original ↗

+ 39 more− less

full release →

~AI code attribution is now preserved by Cursor Blame when formatters like Prettier, Biome, or pre-commit hooks rewrite code.
~File deletions are tracked by Cursor Blame to reflect the full lifecycle of code.
~Clipboard operations on Linux now work with Wayland (wl-copy) and X11 (xclip) for better compatibility across desktop environments.
~Agent sessions are now saved as JSONL transcripts. Headless mode also writes transcripts, making it easier to review and debug non-interactive runs.
~CLI now stores conversation transcripts that the agent can use as context.
~Unified domain allowlisting for WebSearch and WebFetch.
~Known-safe URLs (e.g. Cursor docs) are auto-approved without permission prompts.
~When an MCP server's credentials expire mid-session, the agent can now re-authenticate on demand instead of failing silently.

13 more changed on the release page →

↑Reduced padding, removed the sandbox lock icon from shell prompts, and tightened up visual spacing throughout the CLI interface.

✓Auto-updates now run in the background for faster time-to-interactive.
✓Reduced boot time.
✓Automatically retries connection on transport errors and stalls, with checkpoint-aware state recovery.
✓Background subagent resumes are now queued instead of interrupting running subagents.
✓403 errors no longer trigger logout.
✓Fixed hanging headless mode when workspace trust is required; -force now implicitly trusts the workspace.
✓ApprovalMode: "unrestricted" is now respected without requiring -force.
✓Fixed abort restore so when no checkpoint exists, state rolls back cleanly and user text is preserved.

9 more fixed on the release page →

feb 18feb 18

2026-02-17

v2.1.45claude

4 new 4 improved 7 fixed

+Added support for Claude Sonnet 4.6
+Added support for reading enabledPlugins and `extraKnownMarketplaces` from `--add-dir` directories
+Added spinnerTipsOverride setting to customize spinner tips — configure `tips` with an array of custom tip strings, and optionally set `excludeDefault: true` to show only your custom tips instead of the built-in ones

original ↗

+ 12 more− less

full release →

+Added SDKRateLimitInfo and `SDKRateLimitEvent` types to the SDK, enabling consumers to receive rate limit status updates including utilization, reset times, and overage information

↑Improved startup performance by removing eager loading of session history for stats caching
↑Improved memory usage for shell commands that produce large output — RSS no longer grows unboundedly with command output size
↑Improved collapsed read/search groups to show the current file or search pattern being processed beneath the summary line while active
↑VSCodeImproved permission destination choice (project/user/session) to persist across sessions

✓Fixed Agent Teams teammates failing on Bedrock, Vertex, and Foundry by propagating API provider environment variables to tmux-spawned processes (anthropics/claude-code#23561)
✓Fixed sandbox "operation not permitted" errors when writing temporary files on macOS by using the correct per-user temp directory (anthropics/claude-code#21654)
✓Fixed Task tool (backgrounded agents) crashing with a `ReferenceError` on completion (anthropics/claude-code#22087)
✓Fixed autocomplete suggestions not being accepted on Enter when images are pasted in the input
✓Fixed skills invoked by subagents incorrectly appearing in main session context after compaction
✓Fixed excessive .claude.json.backup files accumulating on every startup
✓Fixed plugin-provided commands, agents, and hooks not being available immediately after installation without requiring a restart

feb 17feb 17

Plugins, Sandbox Access Controls, and Async Subagents

cursor

3 changed 16 improved 7 fixed

~User config only: restricted to domains in your sandbox.json
~User config with defaults: restricted to your allowlist plus Cursor's built-in defaults
~Allow all: unrestricted network access within the sandbox

original ↗

+ 23 more− less

full release →

↑Agents can now search past conversations and use chat history as context.
↑The Cursor CLI agent can handle the sudo password prompt inline for commands that require elevated privileges.
↑Common operations like git clone, npm install, and pip install now work out of the box in the agent sandbox. You can extend or override these defaults per-project.
↑When the agent is in Plan mode, you can now choose "Build in Cloud" to hand off plan execution to a Cloud Agent while you continue working locally or close your laptop.
↑Toggle inline diffs on or off in your settings. By default, diffs are shown only in the review panel.
↑Renamed "Duplicate Chat" to "Fork Chat" in the three-dot menu of a chat message.
↑Improved permission request flow for subagents.
↑Improved the performance of very long chats.

8 more improved on the release page →

✓Fixed bug where some terminal tool calls caused degraded performance.
✓Fixed keybinding behavior for Cmd+Opt Left/Right for tab navigation.
✓Fixed an auto-run mode switching bug.
✓Fixed errors when creating project rules with no workspace open.
✓The global ignore list is now empty by default to fix sandboxing issues. Existing ignore patterns still work the same.
✓Stopping the parent agent will always stop the child subagents.
✓Enforce read-only sandbox for Ask mode even with "Run everything" enabled.

feb 17feb 17

2026-02-16

v2.1.44claude

✓Fixed ENAMETOOLONG errors for deeply-nested directory paths
✓Fixed auth refresh errors

original ↗

feb 16feb 16

2026-02-13

v2.1.42claude

1 new 2 improved 2 fixed

+Added one-time Opus 4.6 effort callout for eligible users
↑Improved startup performance by deferring Zod schema construction
↑Improved prompt cache hit rates by moving date out of system prompt

original ↗

+ 2 more− less

full release →

✓Fixed /resume showing interrupt messages as session titles
✓Fixed image dimension limit errors to suggest /compact

feb 13feb 13

v2.1.41claude

4 new 3 improved 15 fixed

+Added guard against launching Claude Code inside another Claude Code session
+Added speed attribute to OTel events and trace spans for fast mode visibility
+Added claude auth login, `claude auth status`, and `claude auth logout` CLI subcommands

original ↗

+ 19 more− less

full release →

+Added Windows ARM64 (win32-arm64) native binary support

↑Improved model error messages for Bedrock/Vertex/Foundry users with fallback suggestions
↑Improved /rename to auto-generate session name from conversation context when called without arguments
↑Improved narrow terminal layout for prompt footer

✓Fixed Agent Teams using wrong model identifier for Bedrock, Vertex, and Foundry customers
✓Fixed a crash when MCP tools return image content during streaming
✓Fixed /resume session previews showing raw XML tags instead of readable command names
✓Fixed plugin browse showing misleading "Space to Toggle" hint for already-installed plugins
✓Fixed hook blocking errors (exit code 2) not showing stderr to the user
✓Fixed file resolution failing for @-mentions with anchor fragments (e.g., `@README.md#installation`)
✓Fixed FileReadTool blocking the process on FIFOs, `/dev/stdin`, and large files
✓Fixed background task notifications not being delivered in streaming Agent SDK mode

7 more fixed on the release page →

feb 13feb 13

2026-02-12

Long-running Agents in Research Preview

cursor

~Cursor can now work autonomously over longer horizons to complete larger, more complex tasks.

original ↗

+ read the post− less

Cursor can now work autonomously over longer horizons to complete larger, more complex tasks. Long-running agents plan first and finish more difficult work without human intervention.

In research preview and internal testing, long-running agents completed work that was previously too hard for regular agents. This led to larger, more complete PRs with fewer obvious follow-ups.

Cursor's long-running agent is now available at cursor.com/agents for Ultra, Teams, and Enterprise plans.

2026-02-10

v2.1.39claude

1 improved 4 fixed

↑Improved terminal rendering performance
✓Fixed fatal errors being swallowed instead of displayed
✓Fixed process hanging after session close

original ↗

+ 2 more− less

full release →

✓Fixed character loss at terminal screen boundary
✓Fixed blank lines in verbose transcript view

feb 10feb 10

v2.1.38claude

1 changed 1 improved 5 fixed

~Blocked writes to .claude/skills directory in sandbox mode
↑Improved heredoc delimiter parsing to prevent command smuggling
✓Fixed VS Code terminal scroll-to-top regression introduced in 2.1.37

original ↗

+ 4 more− less

full release →

✓Fixed Tab key queueing slash commands instead of autocompleting
✓Fixed bash permission matching for commands using environment variable wrappers
✓Fixed text between tool uses disappearing when not using streaming
✓Fixed duplicate sessions when resuming in VS Code extension

feb 10feb 10

2026-02-07

v2.1.37claude

✓Fixed an issue where /fast was not immediately available after enabling /extra-usage

original ↗

feb 7feb 7

v2.1.36claude

~Fast mode is now available for Opus 4.6. Learn more at https://code.claude.com/docs/en/fast-mode

original ↗

feb 7feb 7

2026-02-06

v2.1.34claude

✓Fixed a crash when agent teams setting changed between renders
✓Fixed a bug where commands excluded from sandboxing (via `sandbox.excludedCommands` or `dangerouslyDisableSandbox`) could bypass the Bash ask permission rule when `autoAllowBashIfSandboxed` was enabled

original ↗

feb 6feb 6

v2.1.33claude

6 new 1 changed 1 improved 8 fixed

+Added TeammateIdle and `TaskCompleted` hook events for multi-agent workflows
+Added support for restricting which sub-agents can be spawned via `Task(agent_type)` syntax in agent "tools" frontmatter
+Added memory frontmatter field support for agents, enabling persistent memory with `user`, `project`, or `local` scope

original ↗

+ 13 more− less

full release →

+Added plugin name to skill descriptions and `/skills` menu for better discoverability
+VSCodeAdded support for remote sessions, allowing OAuth users to browse and resume sessions from claude.ai
+VSCodeAdded git branch and message count to the session picker, with support for searching by branch name

~Errors from invalid managed settings are now surfaced

↑Improved error messages for API connection failures — now shows specific cause (e.g., ECONNREFUSED, SSL errors) instead of generic "Connection error"

✓Fixed agent teammate sessions in tmux to send and receive messages
✓Fixed warnings about agent teams not being available on your current plan
✓Fixed an issue where submitting a new message while the model was in extended thinking would interrupt the thinking phase
✓Fixed an API error that could occur when aborting mid-stream, where whitespace text combined with a thinking block would bypass normalization and produce an invalid request
✓Fixed API proxy compatibility issue where 404 errors on streaming endpoints no longer triggered non-streaming fallback
✓Fixed an issue where proxy settings configured via `settings.json` environment variables were not applied to WebFetch and other HTTP requests on the Node.js build
✓Fixed /resume session picker showing raw XML markup instead of clean titles for sessions started with slash commands
✓VSCodeFixed scroll-to-bottom under-scrolling on initial session load and session switch

feb 6feb 6

2026-02-05

v2.1.32claude

3 new 5 changed 4 fixed

+Added research preview agent teams feature for multi-agent collaboration (token-intensive feature, requires setting CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1)
+Added "Summarize from here" to the message selector, allowing partial conversation summarization.
+VSCodeAdded spinner when loading past conversations list

original ↗

+ 9 more− less

full release →

~Claude Opus 4.6 is now available!
~Claude now automatically records and recalls memories as it works
~Skills defined in .claude/skills/ within additional directories (`--add-dir`) are now loaded automatically.
~Updated --resume to re-use --agent value specified in previous conversation by default.
~Skill character budget now scales with context window (2% of context), so users with larger context windows can see more skill descriptions without truncation

✓Fixed @ file completion showing incorrect relative paths when running from a subdirectory
✓Fixed: Bash tool no longer throws "Bad substitution" errors when heredocs contain JavaScript template literals like `${index + 1}`, which previously interrupted tool execution
✓Fixed Thai/Lao spacing vowels (สระ า, ำ) not rendering correctly in the input field
✓VSCodeFixed slash commands incorrectly being executed when pressing Enter with preceding text in the input field

feb 5feb 5

2026-02-04

v2.1.31claude

2 new 1 removed 3 improved 5 fixed

+Added session resume hint on exit, showing how to continue your conversation later
+Added support for full-width (zenkaku) space input from Japanese IME in checkbox selection
−Removed misleading Anthropic API pricing from model selector for third-party provider (Bedrock, Vertex, Foundry) users

original ↗

+ 8 more− less

full release →

↑Improved system prompts to more clearly guide the model toward using dedicated tools (Read, Edit, Glob, Grep) instead of bash equivalents (`cat`, `sed`, `grep`, `find`), reducing unnecessary bash command usage
↑Improved PDF and request size error messages to show actual limits (100 pages, 20MB)
↑Reduced layout jitter in the terminal when the spinner appears and disappears during streaming

✓Fixed PDF too large errors permanently locking up sessions, requiring users to start a new conversation
✓Fixed bash commands incorrectly reporting failure with "Read-only file system" errors when sandbox mode was enabled
✓Fixed a crash that made sessions unusable after entering plan mode when project config in `~/.claude.json` was missing default fields
✓Fixed temperatureOverride being silently ignored in the streaming API path, causing all streaming requests to use the default temperature (1) regardless of the configured override
✓Fixed LSP shutdown/exit compatibility with strict language servers that reject null params

feb 4feb 4

2026-02-03

v2.1.30claude

7 new 1 changed 2 improved 9 fixed

+Added pages parameter to the Read tool for PDFs, allowing specific page ranges to be read (e.g., `pages: "1-5"`). Large PDFs (>10 pages) now return a lightweight reference when `@` mentioned instead of being inlined into context.
+Added pre-configured OAuth client credentials for MCP servers that don't support Dynamic Client Registration (e.g., Slack). Use `--client-id` and `--client-secret` with `claude mcp add`.
+Added /debug for Claude to help troubleshoot the current session

original ↗

+ 16 more− less

full release →

+Added support for additional git log and `git show` flags in read-only mode (e.g., `--topo-order`, `--cherry-pick`, `--format`, `--raw`)
+Added token count, tool uses, and duration metrics to Task tool results
+Added reduced motion mode to the config
+VSCodeAdded multiline input support to the "Other" text input in question dialogs (use Shift+Enter for new lines)

~Changed /model to execute immediately instead of being queued

↑Improved memory usage for --resume (68% reduction for users with many sessions) by replacing the session index with lightweight stat-based loading and progressive enrichment
↑Improved TaskStop tool to display the stopped command/task description in the result line instead of a generic "Task stopped" message

✓Fixed phantom "(no content)" text blocks appearing in API conversation history, reducing token waste and potential model confusion
✓Fixed prompt cache not correctly invalidating when tool descriptions or input schemas changed, only when tool names changed
✓Fixed 400 errors that could occur after running `/login` when the conversation contained thinking blocks
✓Fixed a hang when resuming sessions with corrupted transcript files containing `parentUuid` cycles
✓Fixed rate limit message showing incorrect "/upgrade" suggestion for Max 20x users when extra-usage is unavailable
✓Fixed permission dialogs stealing focus while actively typing
✓Fixed subagents not being able to access SDK-provided MCP tools because they were not synced to the shared application state
✓Fixed a regression where Windows users with a `.bashrc` file could not run bash commands

1 more fixed on the release page →

feb 3feb 3

2026-01-31

v2.1.29claude

✓Fixed startup performance issues when resuming sessions that have `saved_hook_context`

original ↗

jan 31jan 31