Released: 2026-09-24
Reading long explanations in a wide or full-screen terminal (the text interface where you type commands) often means your eyes have to sweep back and forth across the entire screen. The new maxProseWidth setting key caps plain-text explanations at a comfortable line width, while tables and code blocks that need the width still expand to full width.
When you take over someone else's project folder, if its config files quietly include telemetry (a mechanism that collects operational data and sends it back to a server) variables, your data could leak without your knowledge. Project and local config files now ignore OpenTelemetry variables that enable exporting, set endpoints, or capture content, such as CLAUDE_CODE_ENABLE_TELEMETRY and OTEL_LOG_*. The startup screen, /status, and claude doctor list the telemetry variables ignored in project config files, as well as the variables that keep telemetry off.
When an enterprise locks down managed-mcp.json tool settings in a managed environment, trying to pair it with claude --chrome to let the AI operate a browser used to be blocked outright. A new managed setting key, allowClaudeInChromeWithManagedMcp, has been added. Once an enterprise admin turns it on, Chrome browser connectivity can coexist with the dedicated tool settings. If the browser feature is blocked by a security policy, the error message shown in the terminal now names this setting as the one that hasn't been enabled.
When a server-deployed application gateway (an intermediary server that forwards network requests) hits a brief database restart or a primary-replica switchover, health checks often report errors immediately. The gateway configuration now adds a store.readiness_grace_seconds parameter. During a brief Postgres database interruption, such as a database failover, the health-check endpoint /readyz can temporarily stay in the ready state.
Several issues that could cause lost thinking context and failed requests during long conversations or when continuing old sessions have been fixed. When resuming a session (the conversational cycle of interaction between you and the AI) with --continue or --resume, earlier messages no longer lose their prior extended thinking due to rewriting and resending. When resuming a conversation, removing previously mentioned built-in tools via --tools no longer drops thinking records either. Running /model, /rename, /artifacts, or other immediately executed slash commands while Claude is working no longer interrupts thinking records. Conversation history containing search results the API cannot decrypt no longer triggers a 400 error, which could occur on turns answered through a third-party gateway. When a redacted_thinking thinking block contains invalid data, Claude Code discards the thinking blocks in the conversation and retries once. Hitting a safety switch while thinking is off but effort is set to maximum no longer errors either.
Several fallback and prompt mechanisms for switching models and summarizing long conversations have been corrected. When a summarization request is rejected during long-conversation compaction, it now automatically retries with a fallback model (the alternative that takes over when the primary option fails). In SDK-hosted sessions such as Claude Desktop, an unanswered usage-purchase prompt no longer switches models on its own and instead simply ends that turn; remote-control clients also see the model-switch notification. Entering a full Fable model ID with /model when the plan requires usage that hasn't been enabled now properly pops up the usage-purchase prompt instead of stalling on an API error. Safety-interception messages from Bedrock and Bedrock Mantle now include the request ID and message ID. Unregistered new models on Vertex AI can still provide web search, and the desktop app now prompts you to pick a different model when it encounters an unknown one.
When signing in with multiple editor windows open on the same computer, credential updates no longer conflict and jam. If a previous process is closed or force-killed partway through refreshing credentials (the sign-in proof that verifies your identity), other windows no longer get stuck for up to a minute on a sign-in error saying another process is refreshing. Sessions that start at the same time and happen to hit a sign-in refresh now also automatically retry fetching the organization's security policy.
When enterprise admins distribute organizational security policies, the settings-parsing logic has become more reliable. If a managed lockdown key such as disableClaudeAiConnectors or allowManagedPermissionRulesOnly is given a mistyped boolean value, the security lockdown is still enforced and the key is named at startup. If the permissions, autoMode, worktree (a Git feature that lets one project have multiple parallel copies), or attribution blocks contain invalid fields, the remaining valid settings still apply normally. Under any of the following conditions, sandbox.excludedCommands ignores entries from project and local settings:
Managed settings or --settings set allowUnsandboxedCommands: false
Managed settings set allowManagedDomainsOnly: trueIf Windows or WSL administrator policies (HKLM, managed-settings.json) exist but are invalid or unreadable, the user-modifiable HKCU and WSL /etc/claude-code settings will not be applied.
Permission checks at project startup and command execution are now safer and stricter. At startup, CLAUDE.md and rules in macOS /Network are no longer read via repository symlinks (shortcut files that point to another location) using .. or /.vol-style paths, and rule links pointing to /home are no longer listed. Bash permission rules containing a mid-pattern wildcard :* in config files now fully take effect, with a matching-rule notice at startup. Under managed allowManagedPermissionRulesOnly, skills and commands from repository, user, and --add-dir sources, as well as plugin manifests in skill directories, can no longer pre-approve tools on their own via allowed-tools. After a background work unit of a remote session restarts, commands approved on the restored permission prompt no longer run twice, and the display of special characters in authorization prompts has also been improved.
If your extensions happen to use the names anthropic-skills or claude-ai, the loading results may differ from before. The allow rules Skill(anthropic-skills:*) and Skill(claude-ai:*) cover only skills synced from the claude.ai cloud, not plugins or other skills that merely share the name. Local skill folders, command files, and workflow commands under these two namespaces are no longer loaded; same-name plugins still load, but they yield when their name collides with a synced skill. MCP (Model Context Protocol, the protocol that lets AI connect to external tools) servers configured with these names no longer list skills or prompts, though their tools remain usable; to restore the listing, rename the server in the MCP settings.
This default change affects you if you connect to Claude directly through the Anthropic API with telemetry turned off. When using auto mode (the mode where Claude runs tools on its own without asking step by step), tool actions are now judged by the server-side classifier by default. If you don't want this server-side default, set the environment variable CLAUDE_CODE_AUTO_MODE_SERVER=0 to opt out.
Rendering issues in the terminal interface across various window sizes and output situations have been fixed across the board. Launching in fullscreen no longer shows a brief flash of a blank screen. In normal window mode, when running a command deletes the prompt line at the same time and the screen shrinks but is still taller than the terminal height, output no longer shifts out of place or scrambles. In code diffs, when CJK characters or emoji wrap and redraw, the last column no longer leaves behind stale old characters. For projects with a SessionStart hook (a small program triggered when a specific event happens), the example hint text no longer flickers away at startup. In screen-reader mode, in lists inside quotes, and in long lists, when a code block sits at the start of a list item or at the start inside a quote, the blank line at the top of the block is no longer wrongly deleted.
Input handling and tool error reporting in the terminal are now clearer and more accurate. When Bash and PowerShell hit a full disk quota (the storage space limit), they no longer return only the vague Exit code 1 and leave huge files in the temp folder; they now state clearly that space ran out. When tool argument validation fails, all invalid arguments for that call are now listed at once instead of reporting just one each time. When pasting multi-line text during a session (the conversation cycle between you and the AI), if paste mode resets midway, the lines are no longer sent one by one as commands. When a history prompt containing Tab characters is recalled, the cursor no longer stops at a non-final position. The send hint on older Windows Terminal versions now shows ctrl+x ctrl+s. When reading a PDF fails, paths with special characters display properly, and folder names resembling password or invalid no longer mislead about the cause of the error.
Cancel handling and error safeguards when running commands and managing plugins have been corrected. After you cancel, /install-github-app no longer keeps pushing the branch or saving the key; it stops immediately and reports the progress already completed. When running /feedback, /bug, or /share on platforms like Bedrock and Vertex, canceling midway through saving no longer force-writes the report file. When uninstalling a plugin, if the config file still enables that plugin or can't be read, the system stops the uninstall and points out that file, instead of falsely reporting success or deleting settings. If the plugin list can't be read after uninstalling, your saved options and secret information are still kept, and the uninstall tells you this data was preserved. claude remote-control --debug no longer errors out over an unrecognized argument.
Several details of interacting with terminal menus and list windows have been fixed. In fullscreen, the /feedback draft list now shows a scrollbar on mouse hover. In /skills, keys typed right after / no longer move the skills list and instead go into the search box. As you keep typing keywords, the typing cursor no longer jumps back to the list body, avoiding a hidden cursor and misplaced input-method text. Outside fullscreen, the /skills and /mcp lists are no longer two columns narrower. The key hints at the bottom of the agents panel no longer wrap onto two lines with custom long shortcuts, and the collapse hint respects your custom key settings. If you unbind the shortcut for stopping all agents in keybindings.json, the extra separator no longer appears at the bottom of the /tasks dialog.
When using Vim keyboard mode in the terminal input box, several editing commands have been corrected to standard behavior. The >> indent command no longer adds indentation to blank lines, r with a count no longer corrupts text when it exceeds the line length, 2J no longer merges one extra line, and repeated deletes and indents on the last line (like 2dd or 2>>) no longer shift oddly or delete the final line by mistake. After deleting with dd, dj, or dG, or pasting whole lines with p/P, the cursor lands on the first non-blank character; yy yanking no longer moves the cursor unexpectedly, and pressing Esc after an emoji no longer gets stuck inside the character. Using . to repeat x, s, p, d, or c now correctly recognizes a preceding count; when the previous line is wrapped, whole-line editing commands like p/P, o, O, J, >>, and << no longer act on the wrong line. When recalling a prompt from history or pulling it back from the queue, the cursor in normal mode lands precisely on the final character so x can delete it directly.
Continuing very long conversations and managing artifact files is now smoother. Loading speed when resuming a huge session (the conversation cycle between you and the AI) is significantly faster, including long conversation histories that have never been compacted. On Windows, if a conversation can't be resumed because of a file read failure (EBADF), the error message now lists possible causes and troubleshooting suggestions in detail. When publishing an artifact, version labels longer than 60 characters are automatically shortened to avoid publish failures. The /artifacts panel layout has been overhauled: titles line up neatly in a single column, and descriptions that don't fit are omitted as a whole item instead of being cut off mid-word; the list also supports PgUp/PgDn, Home/End, mouse wheel, and click actions.
The built-in claude-api skill has been updated with the latest rules for model billing and managed agent resource management. The skill documentation adds a link explaining refusals in billing: refusing before output counts toward rate limits, while refusing mid-stream is billed as usual. The docs recommend using ant apply to bring Managed Agents managed resources under version control. On the visual side, the ultracode styling of /effort and the input box has been simplified to a clean, plain look, removing the ripples, border decorations, and keyword shimmer; the spinning indicator for dynamic-workflows has also been removed. The mascot Clawd's feet in the terminal startup banner have been repositioned below the corners of its body.
Several issues in the VS Code extension around long conversations and window integration have been fixed. When generating long replies, the panel no longer re-parses the entire text on every character update, improving the laggy stutter where reply content couldn't keep up with the stream. When the message input box grows tall enough to show a scrollbar, the voice input microphone button no longer covers the scrollbar. For a remote-control session (the conversation cycle between you and the AI) started locally, you can now open the local conversation from the Web list item, unless the session is running elsewhere. After the extension host restarts, if the sign-in screen in an editor tab gets stuck, an "unresponsive" notification is shown.
When managing cloud tasks across platforms, several interface operations have been strengthened for viewing associated codebases or confirming recurring schedule times. The GitHub connector settings page now shows the GitHub App's installation and connection status, with steps to connect, install, or reconnect. When a repository is hosted on a non-GitHub Git server, the cloud session's repository menu adds shortcuts to open the repository and open the version comparison page. A running session (the conversation cycle between you and the AI) that already has one repository can now attach repositories from different GitHub owners, including sessions started from Slack, making it easy to compare against upstream projects. Fixed an issue where the next run time of an hourly recurring schedule in a half-hour time zone was off by 30 minutes. Accounts with large numbers of scheduled check reminders now load the Routines page and the sidebar schedule list noticeably faster.
When teams collaborate across departments using Slack's multi-workspace architecture, they often run into situations where the bot fails to auto-join new channels or cannot reply properly. With an org-wide enterprise installation, channel auto-join rules configured for a single workspace are no longer ignored, and Claude joins qualifying new channels as expected. In channels shared between different workspaces of the same organization, Claude no longer stays silent just because its version setting comes from another workspace. If a channel is shared with a workspace that has not installed Claude yet, the prompt message now clearly states which workspace needs the installation, instead of only reporting that the channel cannot be checked. After disconnecting a GitHub organization, removing that authorization from a permission bundle now saves successfully. When a legacy Slack app connects to GitHub Enterprise Server, the progress card now correctly shows the repository name and provides a working Create PR button.
Model switching and message rendering in Slack internal conversations received several improvements. When the older model a thread was using goes offline, the thread migrates directly to an available model, instead of retrying with the slower fallback model on every reply and popping up a notice. File descriptions Claude uploads to Slack now support table formatting, consistent with regular replies. The Agents & tools list no longer mistakes a thread's first message for its name, and later renames sync as well. After a session's cloud work unit restarts, the cost and token (the text unit AI processes and bills by) counts at the bottom of replies no longer inflate severalfold. Plan, table, and detail border cards in replies now default to a wide layout, and newly connected Slack workspaces also follow changes to the current default model.
Added a `maxProseWidth` setting that caps the width of Claude's prose in wide terminals while tables and code blocks keep the full widthAdded a startup notice, and `/status` and `claude doctor` entries, listing telemetry variables in a project's settings files that were ignored or that turned telemetry off
Changed project and local settings to ignore OpenTelemetry variables that turn on export, set its endpoint, or capture content, like `CLAUDE_CODE_ENABLE_TELEMETRY` and `OTEL_LOG_*`Added the `allowClaudeInChromeWithManagedMcp` managed setting to let `claude --chrome` run alongside an exclusive `managed-mcp.json`; the error shown when Chrome is blocked now names itAdded `store.readiness_grace_seconds` to the Claude apps gateway so `/readyz` can stay ready through a short Postgres outage such as a database failoverFixed every request failing with a 400 error in conversations whose history holds web search results the API cannot decrypt (for example, from a turn answered through a third-party gateway)
Fixed more cases of continued or resumed sessions (`--continue`, `--resume`) re-sending earlier messages in a changed form, which could make the API drop Claude's earlier reasoning
Fixed earlier extended thinking being dropped when `/model`, `/rename`, `/artifacts` or another immediate slash command was used while Claude was working
Fixed continued or resumed conversations losing earlier extended thinking when relaunched with a `--tools` list that leaves out a built-in tool offered earlier in the conversation
Fixed sessions failing on every turn with an "Invalid `data` in `redacted_thinking` block" API error; Claude Code now drops the conversation's thinking blocks and retries once
Fixed a failed turn ("Effort 'xhigh' isn't available with thinking turned off") after a safety-related model switch in sessions with thinking off and effort above highFixed compaction failing when the summarization request is refused; it now retries on a fallback model
Fixed an unanswered Fable usage-credits prompt switching models in SDK-hosted sessions such as Claude Desktop; the turn now ends instead, and Remote Control clients now see the model-switch notice
Fixed `/model` with a full Fable model id stopping at an API error instead of opening the usage-credits prompt when the plan needs usage credits that aren't turned on yet
Fixed safeguard block messages on Amazon Bedrock and Bedrock Mantle not showing a request ID; block messages now also show the message ID
Vertex AI: Fixed web search not being offered for models Claude Code doesn't recognize yet, such as newly released ones
Improved the Claude Desktop unknown-model error to suggest switching to a different modelFixed requests failing for up to a minute with an "another Claude Code process is refreshing it" login error after that other process was closed or killed mid-refresh
Fixed sessions started while another Claude Code window was refreshing the sign-in (common with several VS Code windows) not retrying their organization policy fetchFixed managed settings ignoring a mistyped value for boolean lock keys such as `disableClaudeAiConnectors` or `allowManagedPermissionRulesOnly`; the lock now applies and startup names the key
Fixed managed `permissions`, `autoMode`, `worktree` and `attribution` settings being ignored entirely when one nested value was invalid; the rest of the block now still applies
Changed `sandbox.excludedCommands` to ignore project and local settings entries when managed settings or `--settings` set `allowUnsandboxedCommands: false`, or managed `allowManagedDomainsOnly: true`
Changed Windows/WSL managed settings so an admin policy that is present but invalid or unreadable (HKLM, `managed-settings.json`) keeps user-writable HKCU and WSL `/etc/claude-code` from applyingFixed CLAUDE.md and rules being read at startup through a repository symlink reaching macOS's `/Network` via `..` or a `/.vol`-style kernel path, or a rules link to macOS's `/home` being listed
Fixed Bash permission rules with a mid-pattern `:*` being skipped in settings files while `--allowedTools` honored them; they now work from every source, with a startup warning on how they match
Fixed a command approved on a restored permission prompt running twice when a remote session's worker restarted
Fixed repository, user and `--add-dir` skills, commands and skills-directory plugin manifests pre-approving their own tools via `allowed-tools` under managed `allowManagedPermissionRulesOnly`
Improved rendering of unusual Unicode in permission promptsChanged `Skill(anthropic-skills:*)` and `Skill(claude-ai:*)` allow rules to cover only skills synced from claude.ai, not plugins or other skills that merely use such a name
Changed skill folders, command files and workflow commands in the `anthropic-skills` or `claude-ai` namespace to no longer load; a plugin so named still loads but yields name ties to synced skills
Changed MCP servers configured under the name `anthropic-skills` or `claude-ai` to list no skills or prompts (their tools still work); rename the server in your MCP configuration to list them againChanged auto mode to use the server-side classifier by default on a direct Anthropic API connection when telemetry is off (`CLAUDE_CODE_AUTO_MODE_SERVER=0` opts out)Fixed a blank screen flashing before the first frame when starting in fullscreen mode
Fixed garbled, misplaced rows in the non-fullscreen renderer after the screen got shorter while still taller than the terminal, e.g. deleting a prompt line while a shell command streams output
Fixed a stale character left in the last column of a diff when a redrawn line's CJK character or emoji wrapped to the next row
Fixed the prompt's example text flashing and disappearing at startup in projects with a `SessionStart` hook
Fixed screen-reader mode, quoted lists and very long lists dropping the blank lines at the top of a code block that opens a list item, directly or inside a quoteFixed Bash and PowerShell hiding a full disk quota behind "Exit code 1" and leaving large output files in temp
Fixed tool input validation errors naming only an unknown, missing or mistyped parameter when other parameters in the same call were also invalid; those are now listed too
Fixed pasted multi-line text being submitted line by line after the terminal's bracketed paste mode was reset mid-session
Fixed the cursor landing before the end of a prompt recalled from history when the prompt contains a tab
Fixed the send-now hint showing ctrl+enter on terminals that send it as a newline (Windows Terminal before 1.25); it now shows ctrl+x ctrl+s there
Fixed PDF page-read error messages: paths with accented or non-Latin characters now appear readably, and a folder named like "password" or "invalid" can no longer make the error name the wrong causeFixed `claude remote-control --debug` failing with "Unknown argument: --debug", although Remote Control's own eligibility error says to run with `--debug`
Fixed `/install-github-app` saying "cancelled" and then still pushing the branch and saving the API key secret; leaving now stops the remaining steps and reports what was already done
Fixed /feedback, /bug and /share on Bedrock, Vertex and other third-party providers still saving the report file after you cancelled during the save
Fixed plugin uninstall reporting success and deleting the plugin's saved options when its settings file still enabled it or could not be read; it now stops and names the file
Fixed plugin uninstall deleting a plugin's saved options and secrets when the list of installed plugins could not be read after the removal; they are now kept and the uninstall says soAdded a scrollbar to the `/feedback` drafts list in fullscreen mode; it appears while the mouse is over the list
Fixed a key typed right after `/` in `/skills` moving the skill list instead of reaching the search box
Fixed the terminal cursor jumping from the `/skills` search box to the skill list while typing, which could hide the caret and put IME input in the wrong place
Fixed lists with a scrollbar, such as `/skills` and `/mcp`, being two columns narrower outside fullscreen mode, where the scrollbar can never appear
Fixed the agent panel footer wrapping onto two lines with long rebound keys, and its "Esc to collapse" hint ignoring a rebound collapse key
Fixed a doubled ` · ` separator in the `/tasks` dialog footer when the stop-all-agents shortcut is unbound in `keybindings.json`Fixed vim mode `>>` indenting empty lines, `r` with a count longer than the line changing text, `2J` joining one line too many, and a count on the last line (`2dd`, `2>>`) shifting or deleting it
Fixed vim mode cursor placement: after `dd`, `dj`, `dG` or a whole-line `p`/`P` it lands on the first non-blank, `yy` no longer moves it, and Esc after an emoji no longer leaves it inside the emoji
Fixed vim mode ignoring a count typed before `.` when repeating `x`, `s`, `p`, `d` or `c`, and whole-line `p`/`P`, `o`, `O`, `J`, `>>` and `<<` acting on the wrong line when a line above wraps
Fixed vim mode leaving the cursor past the end of a prompt recalled from history or pulled back from the queue in normal mode, so `x` did nothingImproved the time to resume very large sessions, including ones that were never compacted
Improved the error shown on Windows when a session can't be resumed because its transcript file could not be read (EBADF): it now names possible causes and what to try
Fixed artifact publishes failing when Claude gave the version a label longer than 60 characters; the label is now shortened
Improved `/artifacts`: titles line up in one column, details are dropped whole instead of cut mid-word, and the list supports PgUp/PgDn, Home/End, the mouse wheel and clicksUpdated the `claude-api` skill: pre-output refusal billing now links to the How refusals are billed docs, mid-stream refusals bill at normal rates, and pre-output refusals count against rate limits
Updated the `claude-api` skill to recommend `ant apply` for keeping Managed Agents resources as version-controlled files
Changed the `ultracode` visuals in `/effort` and the prompt input to plain styling (no ripple, border flourish or keyword glimmer) and removed the dynamic-workflows spinner tip
Changed the Clawd mascot's feet in the start-up banner to sit under the corners of his body[VSCode] Fixed long replies falling behind the stream: the panel no longer re-parses the whole reply on every update
[VSCode] Fixed the dictation mic button covering the message input's scrollbar when the input is tall enough to scroll
[VSCode] Fixed Remote Control sessions started on this computer not opening from their Web entry in the session list; they now open the local conversation unless it's running elsewhere
[VSCode] Fixed an editor tab's sign-in screen hanging silently after the extension host restarts; it now shows the "stopped responding" notice too[Cloud sessions] Added Claude GitHub App status to Settings › Connectors › GitHub: whether the app is installed and reachable for your account, plus steps to connect, install or reconnect
[Cloud sessions] Added "Open repository" and "Open compare page" links to the repository menu of a cloud session whose repository is hosted on a Git server other than GitHub
[Cloud sessions] Added attaching a repository from a different GitHub owner, such as a fork's upstream, to a running cloud session that already has one, including sessions started from Slack
[Cloud sessions] Fixed the next run time shown for an hourly routine being 30 minutes off for people in half-hour-offset time zones such as India
[Cloud sessions] Improved how quickly the Routines page and the sidebar's Scheduled list load for accounts whose past sessions scheduled many check-in reminders[Claude Tag] Fixed auto-join channel patterns saved for one workspace in Claude Tag admin settings being ignored on an Enterprise Grid org-wide install; Claude now joins matching new channels
[Claude Tag] Fixed Claude not responding in an Enterprise Grid channel shared between two workspaces of one organization when the channel's Claude Tag version was saved from the other workspace
[Claude Tag] Fixed Claude always replying "Couldn't check this channel just now" in a channel shared with a Grid workspace it isn't added to; the notice now says which workspace needs the app
[Claude Tag] Fixed removing a GitHub organization's grant from an access bundle's Repositories tab in Claude Tag admin settings failing to save after that GitHub organization was disconnected
[Claude Tag] Fixed the earlier Claude in Slack app's progress card for sessions on a GitHub Enterprise Server repository: it now names the repository and offers a working Create PR button[Claude Tag] Fixed Slack threads whose model has been retired falling back to another model on every reply, slower and with a fallback note each time; the thread now moves to a working model
[Claude Tag] Fixed files Claude uploads to Slack not being able to carry a caption containing a table; captions now render with the same formatting as replies
[Claude Tag] Fixed Claude's threads in Slack's Agents & tools view sometimes being listed under their first message instead of their name; later renames now update the list too
[Claude Tag] Fixed the cost and token totals in Claude's reply footer reading many times too high after the session's cloud worker restarted
[Claude Tag] Changed the bordered cards Claude uses in Slack replies for plans, tables and details to render wide by default instead of a narrow width
[Claude Tag] Changed newly connected Slack workspaces to follow the current default model instead of keeping whichever model was the default when they were connected