正文 · AI 翻译
更新内容
- 新增
--safe-mode标志(以及CLAUDE_CODE_SAFE_MODE),可在启动 Claude Code 时禁用所有自定义配置(CLAUDE.md、插件、技能、hooks、MCP 服务器),以便排查问题 - 新增
/cd命令,可将会话移动到新的工作目录,而不会在会话中途破坏提示词缓存 - 新增
disableBundledSkills设置和CLAUDE_CODE_DISABLE_BUNDLED_SKILLS环境变量,用于对模型隐藏内置技能、工作流和内置斜杠命令 - 修复了上/下方向键在长输入行的换行行之间跳转命令历史的问题——现在它们会先逐行移动每个可视行,且历史调取时会从靠近的一侧进入
- 修复了企业托管的 MCP 策略(
allowedMcpServers/deniedMcpServers)在重连时、IDE 输入的配置中、安装后首个会话期间的--mcp-config服务器上、或远程设置加载之前不被执行的问题;同时修复了没有远程设置的组织冷启动缓慢的问题 - 修复了使用 claude.ai 凭据登录的 macOS 用户在每轮对话开始时约 30-50ms 的 UI 卡顿
- 修复了 Windows 上
claude -p在等待斜杠命令/技能扫描时缓慢或看似卡住的问题(2.1.161 中的回归问题) - 修复了恢复会话时恰好发生 OAuth token 刷新导致 Remote Control 卡在“重新连接中”的问题
- 修复了 Windows 上启动时后台 git 命令在无缓存凭据情况下运行时,Git Credential Manager 弹出"Connect to GitHub"窗口的问题
- 修复了使用自定义状态栏的用户看不到底部提示(如"esc to interrupt")的问题
- 修复了重新连接到远程会话时,已过期的权限和对话框提示反复出现的问题(此前该会话的 worker 在等待这些提示期间已崩溃)
- 修复了
claude agents --json遗漏已阻塞和刚派发的后台会话的问题;新增--all以包含已完成的会话,并新增id和state字段 - 修复了在 Windows Terminal 的 WSL 中,从某个 agent 页面返回后 agents 视图残留过期/乱码帧的问题
- 修复了后台 agent 在派发到预热的 worker 时忽略项目级设置
env值(如ANTHROPIC_MODEL)的问题 - 修复了 Windows 上 MCPB 插件缓存被误判失效、导致不必要的重新解压的问题
- 修复了插件
.in_use的 PID 锁文件无限累积的问题;现在每天会清理一次崩溃会话留下的过期标记 - 修复了不受信任的项目设置无需信任确认即可设置 OTEL 客户端证书路径的问题
/workflows现在即使在对话轮次进行中也能立即打开- 改进了
TaskCreate的可靠性:格式错误的输入会自动修复,且未加载工具的验证错误信息中包含 schema - 改进了当你的组织已禁用 API 密钥认证时显示的错误信息,并根据当前使用的 API 密钥来源提供相应指引
- 降低了响应流式输出和加载动画期间的 CPU 占用
- 在 Vertex/Foundry 上恢复了默认的 5 分钟空闲超时,使停滞的流中止而不是无限挂起;设置
API_FORCE_IDLE_TIMEOUT=0可选择退出 - 包含无效条目的远程托管设置现在会应用其余有效策略并显示验证错误,而不是静默丢弃整个配置
- 后台会话现在在 retire→wake 过程中会保留
--ide、--chrome、--bare、--remote-control及其他标志,并且加固了重生状态验证 - 后台会话现在会被告知,在进入 worktree 之前共享检出(shared-checkout)中的编辑被阻止,从而避免在
EnterWorktree之前浪费一次被拒绝的编辑 - “CLAUDE.md 过长”的警告阈值现在会根据模型的上下文窗口大小进行缩放
- Windows 上的自动更新器现在一旦检测到
claude.exe被其他进程占用,就会在当前会话内停止重试 - 改进了斜杠命令菜单中技能标签的颜色对比度
- 对于通过 Apple/Google 计费且未绑定支付方式的订阅用户,领取促销积分时现在会说明在哪里添加支付方式
- 新增了一条提示,在运行多个并发会话时建议使用
claude agents
What's changed
- Added
--safe-modeflag (andCLAUDE_CODE_SAFE_MODE) to start Claude Code with all customizations (CLAUDE.md, plugins, skills, hooks, MCP servers) disabled for troubleshooting - Added
/cdcommand to move a session to a new working directory without breaking the prompt cache mid-session - Added a
disableBundledSkillssetting andCLAUDE_CODE_DISABLE_BUNDLED_SKILLSenvironment variable to hide bundled skills, workflows, and built-in slash commands from the model - Fixed Up/Down arrows jumping to command history past the wrapped rows of a long input line — they now move through each visual row first, and history recall enters at the near edge
- Fixed enterprise managed MCP policies (
allowedMcpServers/deniedMcpServers) not being enforced on reconnect, IDE-typed configs,--mcp-configservers during the first session after install, or before remote settings loaded; also fixed slow cold starts for orgs without remote settings - Fixed a ~30-50ms UI stall at the start of each turn for macOS users logged in with claude.ai credentials
- Fixed
claude -pbeing slow or appearing to hang on Windows while waiting for the slash-command/skill scan (regression in 2.1.161) - Fixed Remote Control getting stuck on "reconnecting" after resuming a session when an OAuth token refresh happened at the same time
- Fixed Git Credential Manager's "Connect to GitHub" popup appearing on Windows at startup when background git commands ran without cached credentials
- Fixed footer hints (e.g. "esc to interrupt") not showing for users with a custom statusline
- Fixed stale permission and dialog prompts reappearing every time you reattached to a remote session whose worker had died while waiting on them
- Fixed
claude agents --jsonomitting blocked and just-dispatched background sessions; added--allto include completed sessions, plus newidandstatefields - Fixed agents view leaving a stale/garbled frame after navigating back from an agent on WSL in Windows Terminal
- Fixed background agents ignoring project-level settings
envvalues (e.g.ANTHROPIC_MODEL) when dispatched onto a pre-warmed worker - Fixed MCPB plugin cache being spuriously invalidated on Windows, causing unnecessary re-extraction
- Fixed plugin
.in_usePID lock files accumulating without bound; stale markers from crashed sessions are now swept once per day - Fixed untrusted project settings being able to set OTEL client-certificate paths without trust confirmation
/workflowsnow opens immediately even while a turn is in progress- Improved
TaskCreatereliability: malformed inputs are repaired automatically and validation errors for unloaded tools include the schema - Improved the error message shown when your organization has disabled API key authentication, with guidance based on where the active API key comes from
- Reduced CPU usage while responses stream and during spinner animations
- Restored a default 5-minute idle timeout on Vertex/Foundry so a stalled stream aborts instead of hanging indefinitely; set
API_FORCE_IDLE_TIMEOUT=0to opt out - Remote-managed settings with an invalid entry now apply their remaining valid policies and surface the validation error, instead of silently dropping the whole payload
- Background sessions now preserve
--ide,--chrome,--bare,--remote-control, and other flags across retire→wake, and respawn state validation was hardened - Background sessions are now told that shared-checkout edits are blocked until they enter a worktree, avoiding a wasted rejected edit before
EnterWorktree - The "CLAUDE.md is too long" warning threshold now scales with the model's context window
- Auto-updater on Windows now stops retrying within a session once
claude.exeis held by another process - Improved color contrast for skill tags in the slash-command menu
- Promo credit claims for Apple/Google-billed subscribers without a payment method now explain where to add one
- Added a tip suggesting
claude agentswhen running multiple concurrent sessions