Skip to content

Claude Code 2.1.292

92changes
  • Add --marketplace <source> to claude plugin install to add and install from a marketplace in one step.
  • Add an effort parameter to the Agent tool to run a sub-agent at a chosen effort level.
  • Speed up rendering of long bulleted or numbered replies.
  • Fix PreToolUse hook approvals and auto mode bypassing the permission prompt for network (UNC) path reads.
  • Fix a managed sandbox read-deny path not ending credential injection from files it covers after it re-points mid-session.

New features

8
More changes (2) Fewer changes

Improvements

20
More changes (19) Fewer changes

Speed up -p and SDK startup

Improved startup of claude -p and SDK sessions: the first turn no longer waits for HTTP and SSE MCP servers to answer resources/list

Keep cleared drafts reachable after Ctrl+C

Improved Ctrl+C draft recovery: a cleared prompt now stays reachable with Up after a slash command or a sent message

Escape system-reminder tags in hook output

Improved hook output handling: <system-reminder> tags written in a hook's output are escaped before they reach Claude

Tolerate minor tool input mistakes

Improved tool input handling: Grep accepts file_path for path, and Write, WebFetch and Read ignore a few stray parameters instead of failing the call

Clarify steps for marketplace name collisions

Improved the steps shown when a marketplace declared in a settings file has a name that looks like an official Anthropic marketplace

Auto-allow interpreter commands with env prefixes

Improved sandbox auto-allow: with strict sandbox mode set in user, managed or --settings settings, an interpreter command with an env var prefix like FOO=bar python3 app.py runs unprompted

List up to 200 artifacts at once

Improved the Artifact tool's listing: Claude now sees how many published artifacts you have and can list up to 200 at once instead of 50

List resumable background agents after a restart

Improved cloud sessions after a restart: Claude is now told which stopped background agents it can resume by id

Clarify Claude in Chrome unreachable message

Improved the Claude in Chrome message in claude.ai cloud sessions when the browser can't be reached: Claude is now told it may continue with alternatives if the user prefers

Improve the /focus tip

Improved the /focus tip: it now invites you to try focus view mid-turn and shows how to switch back

Remember slow legacy MCP servers for 7 days

Improved startup with local (stdio) MCP servers that ignore the newer protocol check: after one slow connect they are remembered for 7 days and connected the older way without the wait

Default to MCP protocol version 2026-07-28

Changed local (stdio) MCP server connections to negotiate protocol version 2026-07-28 by default on every install, including Bedrock, Vertex and Foundry; MCP_PROTOCOL_NEGOTIATION=legacy opts out

Fail plugin tests on hook expect failures

Changed claude plugin test: a failed expect inside a hook the test registered, or a stub answer the engine refuses, now fails the test instead of passing silently

Make usage limit links clickable

Changed usage limit messages to write claude.ai settings links with https:// so terminals and apps can make them clickable

Publish routine artifacts without approval

Changed scheduled and Run now routine runs to publish a new artifact only you can see without asking for approval; artifacts that request connectors or other access still ask

Cap agent names at 256 characters

Changed agent names to allow at most 256 characters: a longer one is rejected, and a skill's or a plugin file's name longer than that is ignored

[Claude Tag] Clarify !status for untagged messages

Improved @Claude !status in a channel to say when Claude has stopped reading its untagged messages, why, and that an @-mention starts it reading again

[Claude Tag] Show origin card on !fork threads

Changed the first message of a Slack thread continued with !fork to a card showing where it came from, the request, and who asked, with a link to the original thread

[Claude Tag] Stop saving spend limits on click-away

Changed the organization-wide and default spend limit boxes on Claude Tag's spend limits page in admin settings to save only when you press Save or Enter, not when you click away

Bug fixes

64

Fix permission bypass for UNC path reads

Fixed PreToolUse hook approvals and auto mode bypassing the permission prompt for file reads from network (UNC) paths

Fix stale grants on re-pointed deny paths

Fixed a managed sandbox read-deny path (and user ones beside it) that appears or re-points mid-session not dropping project grants inside it or ending credential injection from files it covers

Fix sandboxed read of ultrareview uploads

Fixed sandboxed commands being able to read the staged file copies of /ultrareview uploads under ~/.claude/seed-admin

Fix notebook/PDF reads via swapped links

Fixed a notebook or PDF read on macOS and Windows being able to return a file outside what was approved, through a link swapped in mid-read

Fix policy plugin bypass via settings cache

Fixed a tampered on-disk cache of server-managed settings being able to switch off or unseat the built-in policy plugin while the settings fetch failed

Fix auto mode entering when unavailable

Fixed subagent definitions with permissionMode: auto entering auto mode when auto mode is unavailable (disabled by settings, circuit breaker, or a model that doesn't support it)

Wait for managed settings before plugin commands

Fixed claude plugin commands such as marketplace add and install running before an organization's managed settings had loaded on a first run

Fix tool calls skipping permission hooks on restart

Fixed tool calls made while the plugin hooks worker restarts being answered without the plugins' permission hooks

Reject calls from interrupted failing hooks

Fixed a mod's hook that called next(e) and then failed while the turn was interrupted letting the call through; the call is now rejected

Fix tool.check allow skipping user dialogs

Fixed a plugin's tool.check hook answering allow running a tool that requires your answer (a question, a plan approval) without showing its dialog

Fix rm -rf via alternate Windows path spellings

Fixed rm -rf on the 8.3 short name or another alternate Windows spelling of the home folder or a drive not being treated as removing it

More changes (53) Fewer changes

Fix stale allowed-tools after mode exit

Fixed a skill's or slash command's allowed-tools rule coming back in a later turn when you leave auto mode or plan mode partway through that turn

Honor NO_PROXY for Claude Code's own requests

Fixed NO_PROXY being ignored for Claude Code's own API requests (sign-in, policy, feedback, artifacts) when HTTPS_PROXY is set

Fix every-request failure from long tool names

Fixed an MCP tool with a name longer than 128 characters making every request fail; that tool is now left out and an MCP error names it

Wait for background work before exiting -p runs

Fixed one-shot claude -p and Agent SDK runs stopping a background command 5 seconds after the final result, and one-shot claude -p runs dropping a scheduled wakeup; both are now waited for

Restore plan mode on resume

Fixed plan mode not being restored when resuming a session from the claude --resume session picker or with /resume

Fix scheduled tasks lost after resume/branch/clear

Fixed saved scheduled tasks created after /resume, /branch or /clear never firing, and saved tasks ignoring later creates and deletes after two writes to the tasks file milliseconds apart

Fix /loop stopping silently after a restart

Fixed a background session's /loop silently stopping when the session's process restarted (for example after a crash), because its pending wakeup was lost

Report read errors in Grep and Glob

Fixed Grep and Glob reporting no matches when the file or folder they were given could not be read; Claude now retries once or tells you

Report an error for PDF page lists

Fixed the Read tool returning only the first entry, with no error, when a PDF's pages was a list such as "6,9,15"; it now returns an error saying to read each page or range separately

Report large @-mentioned files instead of dropping them

Fixed @-mentioned text files over 256KB being left out silently: Claude is now told the file's size and to read it in portions

Deduplicate usage limit alerts from agents

Fixed the usage limit alert repeating once per background agent when agents failed on a limit that had already stopped the main conversation

Fix empty subagent pane in Remote Control

Fixed Remote Control viewers seeing an empty subagent pane for background subagents in sessions hosted by the desktop app or an IDE

Fix cross-session delivery notice mix-ups

Fixed cross-session delivery notices showing two sessions with similar names as one recipient, and the expiry notice blaming the desktop app when a terminal session let the message lapse

Fix Send now ending a waited-on subagent

Fixed Send now in the desktop app ending the subagent a turn was waiting on when another message was already queued

Fix /bug, /share, /feedback restart and cancel

Fixed /bug, /share and /feedback <text> starting over after Ctrl+O or Ctrl+Z while a report was being sent, and closing as cancelled after it had been sent

Fix /remote-env overwriting the default on Enter

Fixed /remote-env replacing your saved default environment when you pressed Enter right away: the list now opens on your default, and no row has a check mark when no default is in effect

Fix overlapping pastes misread as typed text

Fixed some pasted text reaching Claude as typed text when several pastes overlapped in one prompt

Fix vim mode cursor and motion bugs

Fixed vim mode leaving the cursor past the end of a line, j/k losing their column on shorter lines, and f/t/F/T/;/, jumping to, or deleting up to, a match on another line of the prompt

Fix /add-dir path box line breaks and fast typing

Fixed /add-dir path box letting Shift+Enter or a paste add a line break, and treating fast-typed "tab", "up" or "down" as those keys

Fix dropped input while a footer row is selected

Fixed fast typing, input-method text and decomposed accents being dropped while a prompt footer row was selected, and ! leaving the row selected

Reduce full-screen clears on iTerm2

Fixed fullscreen mode sending a full-screen clear on every window resize and Ctrl+L when iTerm2 is detected, which may be what filled iTerm2's scrollback with stale pages

Fix spurious could-not-be-examined note

Fixed a spurious "could not be examined" note for @-words that name no file when a Read deny rule is set and the working directory is under a symlink

Fix stale instruction-file-not-loaded lines

Fixed "instruction file not loaded" lines going stale or missing after /cd or a permission change, and added a transcript line when a nested one isn't loaded

Block skill invocation via repeated compaction text

Fixed a compaction summary that repeated /name letting Claude invoke a skill that is reserved for the user

Show mod denial reasons on tool rows

Fixed Write, Edit, NotebookEdit and LSP rows, and single Read, Grep and Glob rows, hiding why a mod denied the call: the row now shows the reason

Fix cloud turns stuck as never-ending

Fixed a cloud session showing a turn that never ended when its worker was stopped just as the turn finished

Fix repeated permission prompts in large sessions

Fixed cloud sessions with a large transcript sometimes asking for a permission again after it was approved

Fix queued notifications lost on retry/edit

Fixed scheduled tasks and other queued notifications being lost in cloud sessions when a message was retried or edited while Claude was reading them

Preserve thinking setting across container restarts

Fixed cloud sessions forgetting the thinking setting chosen in the client when the session's container restarted

Fix wrong proxy-blocked artifact message

Fixed Cowork cloud sessions saying a proxy blocked artifacts when Anthropic couldn't confirm the organization's settings

Speed up loading of heavy $.state usage

Fixed plugins whose hooks module makes many $.state calls through one const taking minutes to load or validate

Fix plugin validate misreporting matcher source

Fixed claude plugin validate listing a matcher or state value for a hooks module that the engine reads from elsewhere

Reject plugins with reassigned top-level var

Fixed claude plugin validate listing a $.state value read through a top-level var that was declared again or reassigned; such a module is now refused

Fix infinite guard hook re-run

Fixed a plugin's served $ method restarting the hook origin, which could run a guard hook with a .catch above it again without end

Fix plugin calls skipping hooks during restart

Fixed plugin interface calls made while the plugin hooks worker restarts running without the hooks other plugins put on them

Report denial-after-next(e) hooks as failed

Fixed a mod's config.set, state.set, env.set or agent.spawn hook that denies after calling next(e) being answered as a refusal: the hook is now reported as failed, by name

Ask config.set hook before saving a theme

Fixed /theme, the /config Theme menu and the first-run theme step saving a theme before a plugin's config.set hook was asked

Fix start-up items queued twice on worker swap

Fixed a mod's start-up prompt, command or subagent being queued a second time when the hooks worker was replaced

Fix long deny reasons being ignored

Fixed a plugin's prompt drop or setting deny being ignored when its reason was longer than 4,096 characters

Fix wrong plugin unloaded on $ name clash

Fixed an organization's plugin being unloaded on its own reload, or after another plugin crashed, when it returned a $ name that a user-installed mod had added; the mod is now unloaded instead

Fix tool.call hooks seeing unrepaired params

Fixed plugin tool.call hooks seeing some tool calls before misnamed parameters were repaired; a hook now sees the arguments the tool will run with

Fix guard hook .catch skipped under nested calls

Fixed a mod's guard hook with a .catch being skipped silently for calls another mod's hook makes beneath the guard's own $ call; its .catch is now asked

[Cloud sessions] Fix routines stuck as running

Fixed routine runs occasionally staying listed as running for hours after they had finished

[Cloud sessions] Fix routine edits disabling notifications

Fixed editing or duplicating a routine turning off its push notifications when the routine had no saved notification setting

[Cloud sessions] Attach uncommon image formats as files

Fixed SVG, HEIC, TIFF and other less common image files failing to attach; they now attach as regular files

[Cloud sessions] Remove ineffective Always allow option

Fixed approval prompts offering "Always allow" for connector tools that an organization set to require approval; the choice had no effect

[Remote Control] Accept more file types in the first message

Fixed the first message of a new Remote Control session started from claude.ai/code accepting only images; it now accepts PDFs and other files like later messages

[Claude Tag] Fix held or missed replies during first request

Fixed replies sent in a Slack thread while Claude was still on its first request there being held until that request finished, or missed when sent seconds apart

[Claude Tag] Fix stale model on PR/routine-woken threads

Fixed Slack threads woken only by GitHub pull request activity or a routine staying on their original model after an admin changed the channel or workspace default model

[Claude Tag] Fix wrong spend limit notice on credit exhaustion

Fixed Claude sometimes posting a spend limit notice in Slack when the real cause was that your organization had run out of usage credits

[Claude Tag] Fix session hang after auto-denied prompts

Fixed a session hanging until interrupted when a permission prompt that can't be answered from Slack was denied automatically

[Code Review] Fix queued reviews failing on base deletion

Fixed a queued review failing when the pull request moved to a new base branch and the old one was deleted; the commit is now re-queued for review

[Code Review] Use base branch CLAUDE.md during review

Fixed reviews ignoring a CLAUDE.md's rules when the pull request edits that file; reviews now use its version from the base branch

Change details