Setting CLAUDE_CODE_RESTRICT_PERSONAL_CONFIG now stops your own skills, agents and hooks from widening permissions. In that mode, errors in your user settings file block every prompt and tool call until you fix them. You can set transcriptCacheTtl to 5m to ask for 5-minute prompt caching instead of the default 1 hour. claude mcp serve has a new --session-tunnel mode that connects through a relay. Idle conversations are now compacted shortly before their prompt cache runs out. Plugins can now send notifications through your terminal.
Several features are in this build but not switched on yet. Print mode can show each turn's text result as soon as that turn ends. MCP servers that failed with a retryable error can be retried at the start of the next turn. Images can be re-encoded as smaller WebP files when that saves space. Work on cloud sessions continues, with remote tool requests now pausing while the host computer sleeps. The fleet view can group and pin cloud sessions.
Tool search no longer freezes Claude Code when you have very many tools. A queued message that gets dropped is now handed back to you instead of being discarded. Prompts typed during a left-arrow hand-off to background agents are no longer lost. If a model rejects the 1M-token context option, Claude Code now leaves it out instead of failing every request. Plugin sync from claude.ai no longer removes plugin files that another running session is using. The end-of-turn message no longer includes token budget figures.
Written by our agent from the shipped bundle, not by Anthropic.
In cloud-remote sessions only, claude mcp serve can run over HTTP, dial a session tunnel, read tokens from a descriptor and return tool output objects
ExtensionsA host program can set CLAUDE_CODE_AUTOUPDATER_DISABLED_BY_HOST so plugin auto-updates stay off under the host's version pin
ExtensionsPlugin UI adds a ui.notify operation and Button text from children, and the Grep search now passes a searchOtherAppsData flag
Setting CLAUDE_CODE_RESTRICT_PERSONAL_CONFIG stops your own skills, agents and hooks from widening permissions, and makes key hooks required
Settings & config
CLAUDE_CODE_RESTRICT_PERSONAL_CONFIG mode restricts your own Claude Code filesSetting CLAUDE_CODE_RESTRICT_PERSONAL_CONFIG starts a session that restricts your own Claude Code files, alongside the other restricted modes
Want the reasoning? Read walks the 43 entries that probably matter to you, each one opening to what changed and why.
Read this release → Every row →19 more of these are in What probably matters to you, on page 1.
Claude Code's git checkpoints can now build their file index in a temporary folder and turn off git's split-index feature
In coordinator mode, a refused shell command now comes with the reason it was refused and a hint about how to fix it
For certain MCP servers Claude Code no longer accepts requests to prompt you for input, and cancels any such request it receives
Unclear Which MCP servers are affected is not stated.
For messages from other agents, Slack pings and unclassified senders, all text parts are now joined into one block before Claude reads them
Artifact deletions now always go through Claude Code's shared approval step instead of a separate prompt
Claude Code has a new reason, "MCP server switched off", for blocking an MCP tool, and now tracks reconnecting a server on the next turn
Unclear It is not stated how a server gets switched off, or whether you can do it yourself.
A new guidance mode tells Claude to send file edits and the artifact publish together in one response, without reading the file
When the gateway's Postgres database cannot be written to, the gateway now logs a warning naming the cause and how to fix it
Claude Code now reads every PermissionRequest hook answer as allow, and a PreToolUse answer only when it is set and is not deny
Unclear It is not clear whether this step decides if a tool call is blocked or only records the decision, so it is unknown whether a deny from a hook still stops the call.
When reconnecting to a background session, Claude Code can now stop retrying quietly, and on exit it waits up to 5 seconds for the session to settle
Claude Code includes stricter wording telling Claude to ask you only when it is blocked on a decision that is truly yours to make
Unclear It is not known when the model receives this wording, since the captured stock description did not change.
Claude Code changes how it lists MCP servers that need sign-in and when it reconnects them on the next turn
Unclear What these changes mean in practice for someone using MCP servers is not clear.
Self-hosted runners can now force a refresh that skips the usual wait and re-runs right after one already in progress
Claude Code now tells the model when files staged for a session were not placed
A new note tells Claude that the tools a skill was written for are unavailable and to carry on with the ones it has
Warnings about upstream model lists in proxy settings now apply to every provider, accept wildcards, and flag entries that match no usable model
If a plugin hook on another machine changes a tool call's input, Claude Code now refuses the call instead of running it
Unknown keys in the gateway's Claude Desktop settings are now passed on with a warning, with hints for typos and mis-indented keys
Claude Code no longer reconnects to a saved Cloud gateway when your organization's settings name another gateway or none
In remote sessions, MCP actions that change things and are allowed by policy can now be passed to auto mode's safety check
A wakeup cancellation message, plugin notifications, a model-change reason, a WebP image limit and a new remote switch
Unclear What the new remote switch controls is not stated.
Oversized updatedPermissions answers from SDK programs are now refused or cut, and a waiting permission ask can resolve on its own
Resuming a session now searches abandoned branches of the conversation, and a turn that was cut off is left to the next worker
Claude Code now warns when your deny and ask permission rules are not, or may not be, in force in a cloud session
The warning about the usage-limit cost of resuming a cold cache can now show after a session has been idle too long
With keepsForeignFiles, cloud restore cleanup leaves foreign files in place and warns that deny and ask rules may not be in force
Unclear When this wording is shown instead of the other restart notices is not known.
When you attach to a cloud session, Claude Code now works out whether auto mode is available there instead of copying permissions as they are
The flag-controlled check for destructive MCP tool calls in remote auto mode now also takes the session's MCP servers into account
When a computer Claude is working on goes away, Claude now checks again after 30 seconds, 2 minutes and 5 minutes, and handles a sleeping one separately
Unclear It is not clear whether the remote flag controls all of this or only the timed rechecks.
Setting toolAliases when a session starts now rebuilds the permission settings instead of storing the aliases directly
Unclear It is not clear whether aliased tools are now matched against permission rules differently than before.
claude mcp serve declares extra capabilities in its remote mode#Run in its remote mode, the MCP server started by claude mcp serve now declares more capabilities and accepts session details with each tool call
Unclear It is not clear how the remote mode is turned on or whether a user can reach it today.
The terminal display of quoted text, links and images has been reworked, and mailto: links are now recognised in any letter case
Unclear The exact new appearance of blockquotes and links is not known.
Claude Code now watches for the computer going to sleep and reconnects a remote session after it wakes
Stop in the background tasks dialog now covers more agent states, and a killed mcp_task notifies Claude through the shared path
Unclear It is not clear what stopping an agent that has already finished but is still viewable now does.
Allow and deny rules are now grouped into patterns and withContent lists, which may change which rules match aliased skills
Unclear It is not stated which rule syntax the new pattern entries correspond to or which existing rules would now match differently.
If the plugin hook runner cannot tell where a tool call runs, Claude Code now refuses a hook's rewritten input for that call
Unclear Whether plugin hooks still receive the tool's response in this case is not stated.
The message shown when a request is flagged by safety checks now says "Claude" rather than "Claude Code" in some contexts
In a certain mode, connection errors now show "Unable to connect to API" with the error code, without the internet-connection hint
In a certain mode, Claude Code now skips setting up the memory limit it can place on commands run by tools
Unclear Which mode makes Claude Code skip the memory limit is not settled.
When policy limits cannot be fetched and none are saved, Claude Code now needs one more condition before running without restrictions
Unclear It is not stated what the new check tests, so it is unclear which sessions still run without restrictions.
Agents started by a workflow run can no longer be moved to the background
In a diskless session, OTEL_LOG_RAW_API_BODIES is ignored and some terminal details are left out of usage data
Sessions that keep no files on this machine now refuse to save tool results to disk instead of writing them
If a rewind undoes something Claude wrote to memory, Claude Code now waits and re-checks memory before the next turn
Commands that set simple shell variables are re-checked with the values filled in before being held as possibly writing settings
Claude Code adds a check for a new tip that stays off unless it is switched on remotely
A forked skill agent now refuses to resume, with an error, when the session has no storage to check its permission limits
Claude Code's check for running as a remote session no longer falls back to a second test, so it now depends on CLAUDE_CODE_REMOTE
Unclear It is not known which situations the removed second test covered, so it is unclear who stops being treated as remote.
Data nested beyond a set depth now stops with an error when passed between the sandbox and Claude Code
Claude Code may now skip setting up log export to your own telemetry system under an extra condition
Unclear It is not stated what condition now stops the log exporter being created.
When another session is offline, the model is now told if it went to sleep instead of always being told it failed a liveness check
Deny rules that come from a permission answer are now filtered, and may be dropped or replaced with a warning
Unclear It is not stated which deny rules the filter drops or what it replaces them with.
With --use-anthropic-git-proxy, Claude Code now wipes git settings from several user config folders instead of just one
In diskless sessions, Claude Code now skips creating the tool-results folder and skips some file writes
Unclear What puts a session into diskless mode is not stated.
Invalid session handoff requests now get error messages that name the message or field at fault
Where an item has no content scan, its status can now be "none" rather than always "processing"
Unclear It is not clear which kinds of items this applies to or which setups produce the new "none" status.
The git status snapshot Claude gets at the start can now be taken from a pinned linked worktree instead of the current folder
A diskless session loads no user-level .claude folders or markdown files, and some telemetry storage is skipped
Unclear What makes a session diskless, or a tenant mode active, is not stated.
When a prompt is screened, history and the transcript now record the rewritten text, or nothing if the prompt was dropped
Unclear What causes a prompt to be screened, and whether screening runs for a given user, is not known.
In a session with no disk storage, Claude Code skips listing memory files and refuses to write files for forked skills
Remote-session credentials are no longer written to disk when Claude Code runs inside a tool server's background daemon
Claude in Chrome now matches blocked sites with or without a port, and refuses actions when it cannot tell which site they target
The release notes bundled inside Claude Code now open with a 2.1.292 section, listing an Agent tool effort setting and more
Claude Code now shows a message when it cannot restart, and one version tells you to use /resume to continue your conversation
Changes to the update channel and notification settings in /config now go through one shared check, which rejects unknown values
The confirmation shown before removing a marketplace now counts and lists every installed plugin from it that will be uninstalled
When a request is refused with HTTP 403, Claude Code now keeps the name of the refused resource from the response
Claude Code reads a plugin's name from its plugin.json file, and cloning a plugin marketplace now also passes a path
On exit, Claude Code can now warn that still-running background sessions will stop in about a minute unless you run claude agents
Removing a marketplace or ignored-name entries from the plugin errors tab now needs a yes/no confirmation instead of happening at once
Claude now tells you an MCP server needs signing in only when your request depends on it or you ask, and otherwise carries on
Claude Code now trims stored copies of cut-off MCP tool results and records when a rule allows a destructive MCP call in auto mode
With a gateway provider and no Haiku-class model set, hooks use the main model and failed side requests retry on the default
Downloading a message's attached files can now be stopped partway, and nothing is written to disk once it is stopped
After marketplace add succeeds, Claude Code can now print an extra warning line under the usual confirmation lines
Unclear What the warning says and when it appears are not known.
The gateway sign-in screen can now say its address came from your settings file, and Esc now leads to other ways to log in
Titles, tool names and action descriptions in the cloud session list are now cleaned of control characters and shortened
Self-hosted runner help now says session state stays on disk when cleanup is off, and marks warn as the default confinement mode
Typing over the session feedback survey now abandons it, and the survey records more about how it was answered or undone
Unclear Something not identified decides whether the survey can stay open, so whether typing over it always dismisses it is not settled.
If a Remote Control environment refuses a cloud session, the error now tells you to pick a cloud environment with /remote-env
A new message tells you which attached names were links and so were not sent
On macOS, Claude Code's file listing now leaves out other apps' data folders under ~/Library and stops counting files early from your home folder
Cloud sessions that fail to start now show a reason, and paused sessions waiting on you show as blocked instead of working
Permission prompts and read checks for artifact tools no longer depend on an earlier condition and now always apply
In headless mode, some sessions now refuse to compile schemas, and content lists over a size cap are rejected as too many items
Unclear It is not clear which sessions refuse to compile schemas.
Cloning a plugin marketplace now fetches only the git submodules its plugins point to, and falls back to fetching all when unsure
Claude's Grep tool now understands -l and -c as output modes and ignores -r, noting each change, instead of failing
After switching models, /model now checks whether saving it as your default worked and words its confirmation to match
A new note tells Claude that titles and page text in web results are untrusted and must not be followed as instructions
When Bedrock rejects a model for on-demand use, the gateway now logs that it needs an inference profile and suggests mapping to one
Checking a plugin now advises authors to add an install command to its README when one is missing
Unclear It is not stated which command produces this advice, though claude plugin validate is the likely one.
Removing the built-in plugin marketplace now asks first and keeps its plugins, and plugin checks gained extra tests
When moving a session to the background is cancelled, Claude Code now says why: unsent text in the input box, or a stopped turn
NotebookEdit and file writes now compare what Claude last saw with the file itself, instead of comparing timestamps
Async hooks can now print a multi-line JSON answer, and unreadable JSON lines are logged as errors
Git's stuck-clone check now watches server-side progress before data arrives, has a time cap, and names the cause when it stops git
Installing, enabling or disabling a plugin now warns if the settings file it wrote to will not load and is being ignored
claude plugin validate runs an extra round of checks#claude plugin validate now runs a further round of checks and shows the results in both its text and JSON output
Unclear What the extra checks look for is not known.
Sharing, watching or uploading an artifact always uses the approval card, and publish refuses if an earlier edit in the same reply failed
Remote session views now skip the start-up system message, and a failed login refresh after a rejection is handled cleanly
Published verbatim by Anthropic for v2.1.295. Text is unmodified from the upstream changelog. Everything else on this page came out of the bundle instead, which is why the two lists don't match.
Of these 143 bullets, 35 name something an entry on this page also names, 32 name something no entry here does, and 76 name nothing specific enough to line up either way. The pairings are made on names both sides wrote down, a flag or a setting or a slash command, so read one as probably the same thing rather than as a fact, and read the middle number as candidates rather than as a miss count.
onFailure: "block" for command and HTTP hooks: a hook that can't start, times out, or exits with an unexpected code blocks the action instead of letting it through
No entry names this /copy picker, so a drafted message copies without its > markers
Probably copy-picker-now-lists-blockquotes-as-well-as-code-blocks claude plugin install, enable, disable and marketplace add when the settings file they write to does not load
Probably release-notes-bundle-for-21292, marketplace-add-warns-after-success claude -p run is waiting for when it stays open after its last turn
Probably print-mode-text-results-printed-at-each-turn-end timeouts.upstream_ttfb_ms on the Claude apps gateway's Bedrock, Vertex, Foundry and other cloud upstreams: a value you set now limits how long a stream may take to start there, after which it fails over or gets a 502
Probably gateway-upstream-ttfb-timeout-now-only-applies-when-operator, gateway-notes-upstream-ttfb-ms-now-applies-to-non-anthropic ← is waiting for the current tool to finish
No entry names this models list to every Claude apps gateway upstream: only the listed models are sent there, on failover too, and one * in an entry is a wildcard
Nothing to match on forceLoginMethod: "gateway" and forceLoginGatewayUrl in your own user settings on machines with no managed settings, so /login opens on that Claude apps gateway
No entry names this claude plugin validate when a plugin's README has no install line: it prints the line to paste and never changes the exit code, even with --strict
Probably claude-plugin-validate-adds-a-further-check-pass upstream_request_id to the Claude apps gateway's inference audit event: the request ID from Amazon Bedrock, the Anthropic API or another upstream, for support cases
Probably telemetry-additions-api-message-id-text-runs-joined-upstr $.ui.notify for mods: raises a native notification through your own notification setting and says which channel sent it
Probably plugin-uinotify-channel-and-tool-search-aware-description Button: strings and Text, so a row of a list is one pressable with a chip or a dim detail inside
No entry names this CLAUDE_CODE_RETRY_WATCHDOG_MAX_WAIT_MS to limit how long unattended retry mode (CLAUDE_CODE_RETRY_WATCHDOG) waits out 429 and 529 errors
Probably retry-watchdog-skip-can-be-overridden-by-claude-code-retry-w request-id header to the Claude apps gateway's successful inference responses, so request_id in Claude Code telemetry matches the gateway's audit log
Probably gateway-forwards-upstream-request-id, telemetry-additions-api-message-id-text-runs-joined-upstr [1m] model when a gateway, Bedrock, Vertex or Foundry refuses the context-1m beta; Claude Code now resends without it
No entry names this claude -p text output dropping earlier responses when background work started another turn; each turn's response now prints when the turn ends
Probably print-mode-text-results-printed-at-each-turn-end command_description instead of description
Probably bash-tool-input-command-description-is-accepted-as-descript host:80) to plain http:// pages on that host
No entry names this /plugin's Errors tab removing a marketplace that failed to load, and uninstalling its plugins, on Enter without asking first
Probably plugin-name-read-from-pluginjson-marketplace-clone-gains-p, plugin-errors-view-asks-for-confirmation-before-removing-mar, plugin-addressed-mcp-requestresponse-metadata-gated-behind claude plugin marketplace add reporting success for a marketplace whose name no plugin can be installed under; such an add is now refused
No entry names this CLAUDE_AUTO_BACKGROUND_TASKS moving a subagent to the background while an edit or shell command waited behind it, which started that call before the subagent finished
No entry names this claude remote-control sessions right after a phone message, without the /config toggle, or when started from inside Claude Code
Probably settings-panel-auto-update-channel-and-notification-changes, config-rows-commit-through-a-shared-writer-output-style-an #92661PushNotification reports "Remote Control inactive" (no_transport) in sessions served by `claude remote-control` Open
#99781PushNotification reports "Remote Control inactive" in sessions started with claude rc Closed
/tui exiting without a message, or with a raw system error, when Claude Code could not be started again
No entry names this availableModels not appearing in the /model picker for sessions signed in to a Claude apps gateway; such a model now shows its built-in row in place of a modelPicker row added for the same model
Probably remote-model-now-checks-whether-the-default-save-failed, model-pick-now-goes-through-a-saved-model-row-commit-with-s CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC hiding the Claude apps gateway spend limit in /usage and the status line. The spend limit request goes to the gateway the session is signed in to
Probably spend-limit-and-upsell-hint-text-refactored-into-a-lookup-ta, claude-mcp-serve-gains-an-http-transport-port-and-ses ←; Claude now keeps working in the foreground
Nothing to match on --tools and --restricted not applying to built-in tools that register after launch, and deprecated tool names reaching tools outside the caller's tool set
Probably sdk-spawns-cli-with-flagvalue-form-and-guards-values-st, tools-launch-list-now-withholds-unlisted-tools, new-claude-code-restrict-personal-config-env-var-recognized /loop that was moved to the background with ←; cancelling a pending wakeup now shows a notice
Probably resume-restores-a-pending-loop-wakeup-or-reports-it-as-mis FORCE_COLOR=3, which put color escape codes into the output Claude reads
No entry names this claude agents starting a new background service as it exited when both were stopped with a session open (for example at a reboot), which could delay shutdown and restart an interrupted session
Probably background-attach-prints-exit-notice-about-sessions-stopping #99369Agent view restarts the background daemon during system shutdown; the new daemon holds the shutdown for 90 s Open
worker being shown as "Agent" in auto-mode denial notices and in the activity lists of task detail views
No entry names this TaskStop, which can never be the cause
No entry names this /loop stopping without notice when its next wakeup came due while the session's process was down; the session now says so and Claude is told
Probably resume-restores-a-pending-loop-wakeup-or-reports-it-as-mis /advisor dialog showing a checkmark on a saved advisor model that is no longer available; it now opens on "No advisor"
No entry names this ~ moving the cursor past the last character of a line, so x after it did nothing, and 3~ running on into the next line
Nothing to match on .catch; such calls are now refused
Nothing to match on claude plugin test passing a session.append hook that removes tool call, tool result or thinking blocks that a real session keeps
No entry names this cat ran without printing it
Nothing to match on constructor or prototype always reading as their default and never reloading the plugin when edited
Nothing to match on /reload-plugins or session start was reading the mod's files
No entry names this /model, /fast and /output-style saving their setting without asking a plugin's config.set hook
Probably fast-mode-toggle-wrapped-for-remote-sessions-with-pre-switch, remote-model-now-checks-whether-the-default-save-failed, model-pick-now-goes-through-a-saved-model-row-commit-with-s claude mcp serve background Bash results not naming the output file, and its tool description promising a notification that never arrives
Probably mcp-tool-server-registers-extra-capabilities-when-serving-in, claude-mcp-serve-gains-an-http-transport-port-and-ses, mcp-serve-session-tunnel-and-tool-output-protocol #99934[BUG] `claude mcp serve`: Bash run_in_background output is unreachable since TaskOutput was removed (2.1.277) Open
CLAUDE_ENV_FILE not reaching the Bash tool after an in-app /resume or /branch
Probably restart-failure-messages #98933[BUG] CLAUDE_ENV_FILE env lost after in-app /resume — hook writes to startup session dir, Bash reads resumed dir (still on 2.1.287; #40391, #24775 auto-closed) Open
allowed-tools and effort being dropped when the Skill tool finished before the response stream ended, which denied the skill's Bash commands in -p runs
Probably new-claude-code-restrict-personal-config-mode-limits-what-a #99353[BUG] Skill allowed-tools rule is dropped when the Skill tool finishes before the response stream ends (2.1.289) Open
--plugin-dir plugin's folder in -p and SDK sessions; they are now written only where a mod is being developed
No entry names this /model saying a Max effort pick was saved as your default for new sessions; Max applies to the current session only
Probably remote-model-now-checks-whether-the-default-save-failed, model-pick-now-goes-through-a-saved-model-row-commit-with-s #99350[BUG] Clarify /model confirmation wording: max effort applies only to the current session Closed
pages parameter instead of treating it as omitted
Nothing to match on #98651[BUG] Read: `pages: ""` on a non-PDF file fails validation, and PreToolUse hooks can't work around it (still on 2.1.286) Closed
plugin install and uninstall at project or local scope on Windows missing the install record, or adding a second one, when only the drive letter's case differs
Probably release-notes-bundle-for-21292, plugin-install-lookup-refactored-into-a-shared-helper, plugin-install-records-matched-across-path-spellings-on-wind #92121Project-scope plugin installs are auto-created, never updated, and unreachable via --scope project (Windows drive-case duplicates) Open
↑ or Esc just after ←: prompt history now keeps it
Probably gateway-login-developer-sourced-url-and-esc-wording prompt.submit hook rewrote or dropped still being saved as typed to prompt history and to the transcript's queued-prompt records
No entry names this claude_code.auth OpenTelemetry login event not being emitted on Claude apps gateway sign-in. The event is exported when OpenTelemetry is configured on the machine or the session is already signed in to the gateway
No entry names this ~/.claude/seed-admin by an interrupted /ultrareview upload never being removed by the retention cleanup
No entry names this disableAutoMode was removed from settings
No entry names this /rewind being lost, and the removed turns coming back, when the session was moved to the background or resumed after being killed
Probably spend-limit-and-upsell-hint-text-refactored-into-a-lookup-ta session.receive hook asked for permission before passing a message on
No entry names this setMcpServers() as your own
No entry names this /config saving before a plugin's config.set hook was asked
Probably settings-panel-auto-update-channel-and-notification-changes, config-rows-commit-through-a-shared-writer-output-style-an /context and large file reads use: the gateway now gets them from AWS's CountTokens API instead of a one-token model request. Grant bedrock:CountTokens to use it
Probably bedrock-count-tokens-now-uses-aws-counttokens-with-backoff mcp__server__tool identifier
Probably auto-mode-resolves-toolaliases-for-mcp-tools-behind-tengu-so models: in Amazon Bedrock regions outside AWS's US geography. AWS always refused those requests; the gateway now tries the model in your own region, and a refusal names the model to add
Nothing to match on rate_limit_event usage-limit warnings to say whether the account has extra usage turned on
No entry names this plugin-authoring skill: it no longer tells a session with no terminal to run terminal commands, and explains sharing a mod only when asked
No entry names this -l, -c or -r flag now runs instead of failing the call
Nothing to match on claude plugin validate and plugin loading: when a hooks module is refused over a rebound top-level var, the error now names the line that rebinds it, the cause, and a fix
Probably claude-plugin-validate-adds-a-further-check-pass scriptPath refusal: it now says to pass the script inline via script, the route that works in sessions without a Read tool
No entry names this skills field, each once; a subagent with the Skill tool can still invoke the rest
Nothing to match on /loop wakeup alone, so pressing it twice detaches and the loop keeps running; press Esc to stop it
Probably resume-restores-a-pending-loop-wakeup-or-reports-it-as-mis claude agents stopped with its background service (macOS, or Linux without the service installed): running sessions now stop in about a minute unless it is run again, and a notice says so
Probably background-attach-prints-exit-notice-about-sessions-stopping MCP_PROTOCOL_NEGOTIATION=legacy opts out
No entry names this desktop policy key its bundled Claude Desktop schema doesn't know, so new Desktop settings need no gateway upgrade
Nothing to match on ws) MCP servers: a message over 16 MiB is no longer parsed and closes the connection, the limit the other transports already have
Nothing to match on CCR_AUTO_MODE_ALLOW, CCR_AUTO_MODE_ENVIRONMENT and CCR_AUTO_MODE_SOFT_DENY into a session's environment
Probably runner-spawns-child-with-ccr-auto-mode-env-vars-cleared CLAUDE_RUNNER_FETCH_SERVER_PROGRESS_CAP_MS tunes or turns off the wait
Probably git-fetch-gets-a-server-progress-stall-cap-env-var @Claude !restart had already confirmed the restart
No entry names this agent hook evaluations taking much longer at xhigh and max effort
Nothing to match on A model matched these bullets to the GitHub issues they fix, so a link can be wrong.
1 added and 1 removed, of 218 lines, about 22 words, in the prompt 14 of 27 arms receive. 4 other prompts also changed. The appended system-reminder blocks moved: 1 line added, 1 line removed.
Claude Code, interactive mode
13 prompt changes in this release could not be quoted from the build, so no entry on this page describes them.
568 documentation changes were recorded within 24 hours either side of this release, nearest first. The closest 12 are below. They're here because they happened near this release in time. That's not a claim that this release caused the edit, or that the page documents anything in it.
The 78 literal strings found in the bundle, with the number of entries that name each one. Picking one searches for it. A name is here because this build's code mentions it, which is not the same as it working or being finished.
What's wrong with this entry?