# Claude Code v2.1.312

> Claude Code v2.1.312, released 27 Oct 2026 (2026-10-27). 265 entries read out of the shipped bundle. Unofficial, and not affiliated with Anthropic.

[Web version](https://changelogs.core-directive.com/v/2.1.312)

You can now set `oidc.user_id_claim` to choose which sign-in token field holds the user ID for the OIDC gateway. It defaults to `sub`. Among the other shipped changes, the prompt asking whether Claude may use a skill now shows the skill's description in a scrollable box. Usage messages now call your short-term limit the "5-hour limit". Programs that embed Claude Code can now mark a question as skipped, answered in chat or dismissed with the `ask_user_question_unanswered_v1` capability. Remote control can now ask a session to start a shell command in the background, and the usual permission checks still apply.

Several features are in this build but are not switched on yet. Cloud sessions can enforce a policy on model and effort changes, but only when a worker policy is set to enforce. Auto mode gains a remote list that can turn off its extra safety checks. A tip about Claude Design shows only if a flag is switched on from Anthropic's servers. A note telling Claude that staged files are still being copied is also off by default.

Among the fixes, agent calls cut off by a container restart now resume, or Claude is told plainly why they cannot. Skills supplied by MCP servers can no longer be run through the Skill tool and now return an "Unknown skill" error. Plan mode is read-only again because its workshop and prototype offers are removed. Signed policy documents are no longer rejected for being older than one already accepted. Mantle is gone from the built-in list of model providers.

## Sections

Each section is its own markdown document of whole entries, in pages of about 40 KB. The whole release in one document is [full.md](https://changelogs.core-directive.com/v/2.1.312/full.md).

- [What probably matters to you](https://changelogs.core-directive.com/v/2.1.312/what-probably-matters-to-you.md): 24 entries
- [New Features](https://changelogs.core-directive.com/v/2.1.312/new-features.md): 1 entry
- [Improvements](https://changelogs.core-directive.com/v/2.1.312/improvements.md): 109 entries, 3 pages
- [Bug Fixes](https://changelogs.core-directive.com/v/2.1.312/bug-fixes.md): 19 entries
- [In Development](https://changelogs.core-directive.com/v/2.1.312/in-development.md): 4 entries
- [Internal Changes](https://changelogs.core-directive.com/v/2.1.312/internal-changes.md): 96 entries, 3 pages
- [Removed](https://changelogs.core-directive.com/v/2.1.312/removed.md): 12 entries

## What probably matters to you

The first 10 of 24, one line each. The section's own pages have every one in full.

- [Remote callers can start background shell tasks with start_shell_task](https://changelogs.core-directive.com/v/2.1.312/e/new-start-shell-task-remote-request-runs-a-background-bash-c.json): A new start_shell_task remote request runs a background Bash command in a session through the normal permission checks; cloud-hosted sessions reject it
- [AskUserQuestion: questions must stand alone, and unanswered questions can be marked](https://changelogs.core-directive.com/v/2.1.312/e/askuserquestion-answers-can-carry-unanswered-annotations.json): AskUserQuestion's instructions now say the user may see only the question dialog, and hosts can mark a question skipped, replied in chat or dismissed
- [Claude in Slack and Teams sessions skip server-managed settings](https://changelogs.core-directive.com/v/2.1.312/e/claude-in-slackteams-sessions-are-excluded-from-server-mana.json): Sessions run by Claude in Slack or Teams now skip server-managed settings, marked by a new `claude_tag` reason; policy limits are decided separately
- [Data-sharing setting can be locked for Claude Science accounts](https://changelogs.core-directive.com/v/2.1.312/e/privacy-settings-dialog-can-be-locked-by-claude-science.json): The privacy dialog and the consumer terms dialog can now lock the data-sharing (Grove) choice for Claude Science accounts, showing it as not available
- [Startup tips can now take their text from a content-link table](https://changelogs.core-directive.com/v/2.1.312/e/tips-content-link-table-and-gated-claude-design-tip.json): Several startup tips, including the /memory tip, now read their text through a content-link table, and the Claude Design tip sits behind a flag
- [Loading an MCP skill can wait for a person to answer the permission prompt](https://changelogs.core-directive.com/v/2.1.312/e/mcp-skill-ask-is-now-held-for-a-person-in-the-held-calls-gat.json): When loading an MCP skill needs permission, Claude Code can now hold the request until a person answers it
- [Workflow agents hand back their report through a tool, and the report can be reviewed](https://changelogs.core-directive.com/v/2.1.312/e/workflow-agents-get-an-opt-in-handback-tool-with-review-and.json): Workflow agents can return their final report through a handback tool, and that output is reviewed beyond auto mode when report review is set
- [The system prompt is now assembled in a new way, with no visible text change from it](https://changelogs.core-directive.com/v/2.1.312/e/system-prompt-builder-reworked-into-factsdecisionsrecord-w.json): Claude Code rebuilt how it assembles its system prompt and can record each section's choice, but the prompt text itself is unchanged by it
- [Server classifier input rewrites are honoured and recorded, with new no-server-check tracking](https://changelogs.core-directive.com/v/2.1.312/e/auto-mode-can-use-an-arbiter-server-check-outside-auto-mod.json): Permission checks now honour and record server classifier input rewrites, can use a server arbiter when auto mode is off, and track missing server checks
- [The OIDC gateway can take the user ID from a claim other than `sub`](https://changelogs.core-directive.com/v/2.1.312/e/oidc-proxy-gains-oidcuser-id-claim.json): A new `oidc.user_id_claim` setting picks which sign-in token field holds the user ID, defaulting to `sub`