claudekit / updates / claude-code-2-1-317
[ PATCH · ]

Claude Code 2.1.317

Adds a check that asks before a web or shell tool call sends your account email address to the network, a way for failed Agent SDK hook callbacks to block the action, and a cost report on the Claude apps gateway admin API. Over 80 changes, including a fix for `claude -p` sharing the session ID and transcript of the cloud or self-hosted runner session it runs inside.

Official announcement →

This article is a summary based on official documentation.

What changed

Claude Code 2.1.317 shipped on November 3, 2026. It adds guards against unintended actions, such as the account email check, blocking on failed Agent SDK hook callbacks and the /schedule connector warning, along with a cost report for the Claude apps gateway. Fixes make up more than half of the release and span MCP servers, sessions and compaction, claude -p, OpenTelemetry exports, agent teams, self-hosted runners and the VS Code extension.

New features

  • Account email check

    A web or shell tool call could send your account email address to the network with no check in between. A new check now asks first.

  • onFailure: "block" for Agent SDK hook callbacks

    A hook callback that failed let the action run. With this option, a callback that fails blocks the action instead.

  • Connector warning in /schedule

    /schedule now warns when a new cloud scheduled task gets connectors you did not ask for, and says how to remove them.

  • Cost report on the Claude apps gateway admin API

    GET /v1/organizations/analytics/cost reads past spend over a date range in total, per developer, per day or per IdP group.

  • Daily record of IdP groups on Claude apps gateways

    Gateways with admin: set now keep a daily record of each developer’s IdP groups for admin.identity_retention_days. The cost report groups by this record.

  • New OTel attributes

    anthropic.plan, anthropic.seat_type, anthropic.provider and anthropic.custom_base_url are now on OTel metrics, events and spans.

  • CLAUDE_CODE_DISABLE_NESTED_SESSION_ID

    claude -p run inside a cloud or self-hosted runner session no longer shares that session’s ID. Set this variable to 1 on a self-hosted runner to have it use that session’s ID and transcript, as in earlier versions.

  • Managed settings notice in Claude Desktop’s Code tab

    A notice now shows when your organization’s remote managed settings haven’t loaded for the session, and another once they do.

  • Auto mode notice for subagent tool calls

    When auto mode’s check had no answer for a subagent’s tool call, nothing was left in the main view. A notice now shows.

  • [Cloud sessions] Link card

    A link an MCP server asked you to visit could go unseen. A link card now lets you open or decline it.

Key improvements

  • Read tool’s “file unchanged since your last Read” stub removed

    Reading an unchanged file again now returns its content instead of the stub.

  • SendMessage to a running background agent

    A message now arrives earlier: the agent’s running command moves to the background, without being stopped.

  • Effort for subagents and teammates on a different model

    Subagents and teammates on a different model from the main conversation followed the session’s effort level. They now use that model’s saved effort level or its default.

  • “Resume this session with” hint at exit

    The hint now repeats the flags you launched with, for example --chrome --debug, ahead of --resume.

  • Startup error when a required Claude apps gateway can’t be reached

    The error now points to the network instead of a sign-in, and says when the gateway’s TLS certificate can’t be verified.

  • Azure credential errors on Microsoft Foundry

    When Azure credentials can’t be loaded, the error now says to check or refresh them, and reports as cloud_credential_error.

  • /design-sync says up front when your design system has moved to Claude

    It failed at upload in this case. It now says so up front and stops before building.

  • Self-hosted runner recovery from server errors

    After a 5xx other than 501 or 529, a runner’s next poll and a session now open new connections instead of reusing the one that failed. After a heartbeat gets no answer, a session opens new connections instead of trying each idle one in turn.

  • Exit with plugin monitors running

    Exit now skips the “Background work is running” prompt for plugin monitors.

  • Sharing offer in the plugin-authoring skill

    The built-in plugin-authoring skill now offers once, after a mod you built works, to help share the mod.

  • Remote Control fork names

    When the app doesn’t set a fork name, a long name now keeps its “(fork)” ending. In sessions started with claude remote-control, a fork of “Name (fork)” is “Name (fork 2)”.

  • [Cloud sessions] Organization environment dialog

    The dialog in admin settings now has the personal dialogs’ short layout: each section opens from a row, and unsaved input asks before closing.

  • [Claude Tag] Artifact link cards in Slack

    Newly pasted artifact links now show a simpler card with an Open button that opens the artifact in Claude, instead of Slack’s side panel.

Bug fixes

MCP, plugins and skills

  • Project and local MCP servers disconnecting after the first turn in some Claude desktop sessions.
  • Stdio MCP servers receiving an empty AWS_BEARER_TOKEN_BEDROCK when their env config sets it to empty — the variable is now unset, so the server can use the machine’s own AWS credentials.
  • Claude saying a working claude.ai connector needed authorization or had failed in cloud sessions when a synced plugin included the same server.
  • Skills synced from claude.ai being lost or downloaded again every ten minutes when they can’t be written to disk.
  • High memory use and slow startup in headless sessions with many plugins that register commands.

Sign-in and networking

  • claude auth status exiting before a slow token refresh was saved, which could sign you out on the next run.
  • A region or network block (HTTP 403) being shown as a sign-in failure — the error now explains the block.
  • API requests failing with “unable to get local issuer certificate” behind some TLS-inspecting proxies after a root certificate renewal.

OpenTelemetry

  • Exports to an https:// endpoint through a proxy never being sent when they started while another export was still in progress.
  • A flood of connection attempts to a proxy that blocks (HTTP 403) exports to an http:// endpoint.
  • Exports hanging indefinitely behind a proxy that never answers — they now time out after 10 seconds, or OTEL_EXPORTER_OTLP_TIMEOUT up to 30 seconds.

Auto mode and permissions

  • Auto mode failing every turn when an LLM gateway set as the base URL answers 403 or 422 to its server-side safety check. A 403 with an apiKeyHelper is unchanged.
  • Auto mode in cloud sessions stopping for approval when the Artifact tool uploaded, wrote to a database or saved a file through the outputs folder.
  • Agent team teammates in tmux or iTerm2 panes starting in auto mode when the lead session was in manual mode.
  • A background session sometimes losing a just-made change of permission mode when it was upgraded or retired.
  • The second of two simultaneous permission prompts showing no file path or command on claude.ai and the mobile apps.
  • Remote Control and SDK hosts offering “don’t ask again” on a mod’s tool calls.

Sessions and compaction

  • “Prompt is too long” after automatic compaction in sessions whose very large CLAUDE.md changed since it was last sent.
  • Very long conversations losing their earlier summary when compaction had to shorten its own request.
  • Resumed sessions telling Claude again about a file change it already knew about.
  • Claude sometimes not using a built-in tool in a resumed session after an earlier call failed because the tool was switched off.
  • Session titles being lost in cloud and SDK sessions when a new message or Stop arrived while the title was generating.
  • claude --from-pr reporting “No conversations found” in projects with many sessions when the linked session was older than the first page.
  • A project’s effortLevel setting being ignored by sessions that an app starts ahead of time and then opens in the project, such as new sessions in the desktop app’s Code tab.
  • Edits to settings files made in the first second or so after launch not taking effect until a settings file changed again.
  • Old transcripts not being cleaned up on machines where some runs skip cleanup, such as runs using --setting-sources.

claude -p

  • Exiting 0 when a full disk or quota lost its transcript or ~/.claude.json writes — it now prints an error and, in text and json output, exits 1.
  • Sharing the session ID of the cloud or self-hosted runner session it runs inside — it wrote into the session’s transcript, and a restart could resume the wrong conversation.
  • Generating a title of its own and sending it to the cloud or self-hosted runner session it runs inside.

Agent teams and background work

  • Deleted agent team inboxes being re-created by a teammate’s late message.
  • A re-created agent teammate being removed from its team by an earlier shutdown approval.
  • Ctrl+C stopping all running background agents when pressed to clear typed text at the prompt.
  • Child processes of a stopped shell command sometimes being sent SIGTERM twice, which could make some programs skip their cleanup.

Scheduled tasks and /team-onboarding

  • Self-paced /loop wakeups sometimes not being scheduled when Claude left a field out of its request.
  • /schedule leaving a cloud scheduled task’s next run time and link out of its results after creating or updating it.
  • Stored scheduled tasks being impossible to list or delete after scheduled tasks were turned off.
  • Claude deleting your most recently updated /team-onboarding guide when asked to delete one without naming which.
  • Claude saying a /team-onboarding guide was deleted when the server refused the delete.

Tools and commands

  • Claude treating partial Grep and Glob results as complete when a search hit its time limit.
  • A PDF’s own metadata being able to make Read and @-mentions under-count its pages.
  • DISABLE_PROMPT_CACHING_OPUS and DISABLE_PROMPT_CACHING_SONNET having no effect on [1m], other-region and undated spellings of the default model.
  • /artifacts and the Artifact tool’s shared list showing no shared artifacts when you have 200 or more of your own.
  • /remote-control answering “Unknown command” instead of saying that Remote Control is turned off by the disableRemoteControl setting.
  • Screen readers speaking a running subagent’s elapsed time and token count on every change in screen reader mode.

Claude apps gateway

  • The gateway hiding Amazon Bedrock’s “Input is too long.” error, which left sessions failing every turn with “upstream rejected the request” instead of compacting.
  • The status line and /usage showing a developer’s own spend limit as 100% used after their group’s pooled budget ran out.

Windows

  • Every tool call being refused on some volumes where the native path lookup fails — the working-directory check now compares the folder’s identity instead.
  • The session scratchpad and other temporary paths growing with your project’s path length in sessions not started by Claude Desktop, which made repositories cloned into them hit the 260-character path limit.

Self-hosted runner

  • Sessions losing recent turns or the whole conversation when restarted while a plan waited for approval.
  • Claude not seeing the interrupted tool call when a session moves to a new runner and the turn is older than CLAUDE_CODE_RESUME_INTERRUPTED_TURN_MAX_AGE_MS.
  • An MCP server that was switched off coming back on after the session moved to a new machine, when the session service had refused to save the switch.

Remote Control

  • The next message to an idle session going unanswered after you restart claude remote-control, when you sent it within a minute of your previous message.

VS Code extension

  • Forked conversations dropping messages — only one of the tool calls Claude ran in parallel was kept, and the last messages a compaction kept were lost, which could leave Claude’s answer without the question it answered.
  • A message being lost when Claude did not finish starting in time — the extension now starts Claude once more before showing the error.
  • The sign-in screen opening when a request is blocked for its region or network.
  • File links in a Claude tab opening on top of the chat instead of in another editor group.
  • Attachment chips, menus, banners and dialogs in the Claude panel staying small when chat.fontSize is raised.
  • ultrathink not raising effort for its turn or showing its rainbow when telemetry is off.

Cloud sessions

  • Adding a network secret for a host that already has one silently replacing it — the Add secret form now refuses and points you to the existing secret.
  • New scheduled tasks ignoring your organization’s default environment — the form now starts on the one an admin chose instead of the Anthropic-hosted default.

Notes

  • New account email check. A web or shell tool call that would send your account email address to the network now asks first.
  • If a self-hosted runner flow relies on claude -p sharing the session ID, set CLAUDE_CODE_DISABLE_NESTED_SESSION_ID to 1 to keep the earlier behavior.
  • Read returns content for unchanged files. Check any automation that expects the “file unchanged since your last Read” stub.
  • Subagents and teammates on a different model use that model’s saved effort level or its default, not the session’s.
  • The claude -p exit code fix has a limit. Full NFS exports may be missed.
  • OpenTelemetry exports now time out after 10 seconds, or OTEL_EXPORTER_OTLP_TIMEOUT up to 30 seconds.
  • This post covers the main items. See the GitHub release notes for the full list.