Claude Code 2.1.233

同事丟來 GitLab 的合併請求網址時，現在可把網址交給 --worktree，也能在 claude agents 檢視裡直接辨認它。工作樹（worktree，Git 為同一專案開出的平行副本）讓你另開一份地方查看修改，不必離開手邊工作。

合併請求會顯示成 !N，因此清單裡能看出它是 GitLab 的項目。

公司透過 apps gateway 連到 Anthropic 上游服務，又想知道費用算在誰身上時，可選擇開啟 forward_user_identity。這個設定會把已登入使用者的身分放進 headers（隨請求附上的資訊欄位）。再交給閘道後方代替你轉送請求的 proxy。

此設定是選用項目；不開啟便不會轉送這份身分資訊。

在 Linux 用 Bash 工具跑建置，若擔心失控的建置拖住工作階段，可選擇設定 CLAUDE_CODE_TOOL_MEMORY_LIMIT。它會替 Bash 指令啟用 memory cgroup（替一組程式設下記憶體界線的 Linux 機制）。

這是選用支援，目的是避免失控的建置讓整個工作階段停住。

同一個 WebFetch 工作階段裡，想調整網址快取保存多久，現在可設定環境變數 CLAUDE_CODE_WEBFETCH_CACHE_TTL_MS。快取（cache，把查過的結果暫存起來直接用）會在工作階段內保存網址查詢結果；這個變數專門調整它的存活時間。

預設值仍是 15 分鐘，沒有因這項設定而改變。

啟動 claude self-hosted-runner 時，現在會先建立工作階段分支，且不必改寫工作樹。工作階段（session，一次持續工作的連線與狀態）一開始少了會擋住 agent 啟動的兩次伺服器往返。

因此，這個 runner 的工作階段啟動時間獲得改善，agent 不再被那兩次往返阻住。

apps gateway 接到 Vertex、Foundry 或 AWS 上 Claude Platform 回傳的 400 或 413 錯誤時，現在會帶出上游服務自己的錯誤訊息。上游（upstream，請求原本要送往的服務端）說明能讓你看到它實際拒絕的理由。

使用 apps gateway 的 auto-compact 時，這項錯誤已修正。auto-compact（自動壓縮，將較早對話整理得更短以騰出內容空間）是 gateway 替工作階段整理內容的流程。

因此，透過 apps gateway 進行自動壓縮時，不再受這項錯誤影響。

準備檢查 plugin 時，claude plugin validate 現在會查看單獨存在的 .claude/skills 資料夾。它會回報其中 frontmatter（檔案開頭給工具讀取的欄位）無法解析的 SKILL.md。

即使 .claude/skills 是裸資料夾而非完整結構，驗證也會納入檢查。

使用螢幕閱讀器調整 /effort 時，選擇器現在會呈現成編號清單，並要求輸入對應數字。螢幕閱讀器（screen reader，把畫面內容讀出聲的輔助工具）需要能循序讀懂每個操作選項與目前選擇。

此外，提示文字和對話框文字不再被裁切，讀到的內容能完整保留。

用 print mode 送出 Claude Code 不認得的 model ID 時，現在 stderr（標準錯誤輸出，程式用來報告問題的文字通道）會寫出 [claude-code:unrecognized_model]。這條診斷訊息告訴你請求確實帶了未識別的模型 ID。

若想不再看到它，可用 modelOverrides 將該模型 ID 對應到已知模型。

你的儲存庫 origin remote 在 gitlab.com 或 bitbucket.org 時，GitHub app 設定提示不再冒出來。origin remote（儲存庫連向遠端來源的位置）既不是 GitHub，就不該叫你設定 GitHub app。

同時，enterprise marketplace 提示會涵蓋不是 GitHub 的內部 Git 主機。

使用 Opus 4.8、Sonnet 5、Fable 5、Mythos 5 或更高版本模型時，TaskCreate、TaskGet、TaskUpdate、TaskList、TodoWrite 這些待辦與任務追蹤工具不再提供。這是既有工具可用範圍的變更，不是新增工具。

要恢復它們，設定環境變數 CLAUDE_CODE_ENABLE_TODO_TOOLS=1。

When a teammate sends a GitLab merge-request URL, you can now give that URL to --worktree and recognize it in the claude agents view. A worktree is a parallel copy of a Git project, so you can inspect changes somewhere separate from current work.

Merge requests appear as !N, making their GitLab identity visible in the list.

If your company connects to Anthropic upstreams through an apps gateway and needs to assign spending to people, it can opt into forward_user_identity. This setting puts the signed-in user's identity into request headers for a proxy behind the gateway.

Leave it off and that identity is not forwarded.

When a build runs through the Bash tool on Linux, you can opt into CLAUDE_CODE_TOOL_MEMORY_LIMIT if a runaway build might stall the session. It enables a memory cgroup, Linux's boundary for the memory available to a group of programs.

The optional limit keeps a runaway build from stalling the session.

Within one WebFetch session, you can now set CLAUDE_CODE_WEBFETCH_CACHE_TTL_MS to choose how long its URL cache lasts. A cache keeps a URL lookup result for direct reuse during the session; this variable controls that lifetime.

The default remains 15 minutes.

When starting claude self-hosted-runner, the session branch is now created without rewriting the worktree. A session is the connection and state for one continuous stretch of work, and its start no longer waits for two server round trips that blocked the agent launch.

The runner's session starts faster because those two round trips no longer block the agent.

When an apps gateway receives a 400 or 413 error from Vertex, Foundry, or Claude Platform on AWS, it now carries the upstream service's own message. An upstream is the service the request was meant to reach, so its message tells you why it rejected the request.

A bug in auto-compact on apps gateway is fixed. Auto-compact condenses earlier conversation to make room for more content in a session.

Auto-compact through apps gateway is no longer affected by this bug.

When checking a plugin, claude plugin validate now examines a bare .claude/skills directory. It reports SKILL.md files whose frontmatter, the fields at a file's beginning for tools to read, cannot be parsed.

Validation includes .claude/skills even when it stands alone rather than inside a full structure.

When using a screen reader to change /effort, the selector now appears as a numbered list and asks for a typed number. A screen reader is an assistive tool that reads screen content aloud and needs an ordered way to convey choices.

Hint text and dialog text are no longer clipped either.

If print mode sends a model ID Claude Code does not recognize, stderr, the program's text channel for reporting problems, now writes [claude-code:unrecognized_model]. The diagnostic line tells you the request went out with an unknown model ID.

Map that ID with modelOverrides to silence the line.

If a repository's origin remote is on gitlab.com or bitbucket.org, the GitHub app setup tip no longer appears. An origin remote is the repository's destination for its remote source, so a non-GitHub destination should not ask for GitHub app setup.

The enterprise marketplace tip now covers internal Git hosts that are not GitHub.

On Opus 4.8, Sonnet 5, Fable 5, Mythos 5, and newer models, the task-tracking tools TaskCreate, TaskGet, TaskUpdate, TaskList, and TodoWrite are no longer available. This changes the availability of existing tools; it does not add a new tool.

Set CLAUDE_CODE_ENABLE_TODO_TOOLS=1 to restore them.