cua-driver 预编译二进制文件 — 相对坐标分支
（托管于 `packages/cua-driver` 下）。
- **macOS**: 已代码签名 + 公证的通用二进制文件 + `QwenCuaDriver.app`
- **Linux**: 未签名（x86_64 + arm64，glibc 2.31 最低版本）
- **Windows**: 未签名（x86_64 + arm64）
启用相对坐标：`CUA_DRIVER_RS_COORDINATE_SPACE=1`
（默认 `0` = 关闭；可选 `CUA_DRIVER_RS_COORDINATE_SCALE=1000`）。
<!-- 使用 cua-driver-rs-v0.7.1 的 .github/release.yml 配置生成的发布说明 -->
## 更新内容
* feat(scheduler): 使重复性 cron/循环任务的过期时间可配置 by @TianYuan1024 in https://github.com/QwenLM/qwen-code/pull/6173
* fix(serve): 遵循已禁用的技能设置 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6223
* fix: 将 vscode-ide-companion 花括号规则与根配置对齐 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6221
* fix(qqbot): 安全加固 — 网关验证、原子状态、脱敏日志 by @Eric-GoodBoy-Tech in https://github.com/QwenLM/qwen-code/pull/6200
* fix(cua-driver): 迁移 Windows 脚本 + 添加 /mcp 批准按钮 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6241
* [codex] 回退 DashScope 的 GLM 标签化思维解析 by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6248
* fix(web-shell): 改进会话恢复和加载反馈 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6220
* feat(mobile-mcp): 引入 mobile-mcp 并支持可选的 0-1000 相对坐标 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6235
* feat(web-shell): 在侧边栏底部显示 qwen-code 版本 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6222
* feat(daemon): 添加会话工件 API by @chiga0 in https://github.com/QwenLM/qwen-code/pull/5895
* feat(web-shell): 在任务面板中以树形结构显示嵌套子代理 by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6239
* 添加无会话工作区记忆遗忘与梦境功能 by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6227
* fix: 避免 vsce 密钥扫描器在正则表达式模式上产生误报 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6247
* fix(web-shell): 对视觉模型选择器选择进行编码并优化分发逻辑 by @AmariahAK in https://github.com/QwenLM/qwen-code/pull/6236
* feat(web-shell): 改进斜杠命令发现功能（更高菜单、分组计数、模糊搜索） by @wenshao in https://github.com/QwenLM/qwen-code/pull/6267
* fix(serve): 优化守护进程 NDJSON 流处理 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6263
* ci(autofix): 恢复沙箱镜像流程 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6261
* fix(qqbot): 优先使用 Markdown 发送、replyMsgId 过期时间 (TTL) 及移除死代码 by @Eric-GoodBoy-Tech in https://github.com/QwenLM/qwen-code/pull/6201
* feat(daemon): 在工作区提供程序状态和 web-shell 模型对话框中暴露 visionModelId by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6262
* fix(triage): 通过批量检测、问题存在性检查和红旗模式强化 PR 门禁 by @pomelo-nwu in https://github.com/QwenLM/qwen-code/pull/5723
* feat(review): 将建议级发现结果路由至可更新的 PR 评论 by @pomelo-nwu in https://github.com/QwenLM/qwen-code/pull/5786
* fix(autofix): 在分支检出前无条件恢复已跟踪文件 (#6281) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6286
* feat(serve): 在守护进程状态 API 中添加 runtime.activity 字段 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6270
* feat(web-shell): 添加由 GET /daemon/status 支持的守护进程状态页面 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6272
* fix(vscode): 失去焦点时保持认证快速输入框打开 by @barry166 in https://github.com/QwenLM/qwen-code/pull/6274
* ci(audio): 明确 macOS 预构建工件后缀 by @barry166 in https://github.com/QwenLM/qwen-code/pull/6275
* fix(cache): 在 Anthropic 提示词缓存命中时保留 side-query 中的 tools 前缀 by @kagura-agent in https://github.com/QwenLM/qwen-code/pull/6225
* feat(web-shell): 添加 MCP 提及和带图标的 @ 引用 by @BZ-D in https://github.com/QwenLM/qwen-code/pull/6279
* fix(core): 为 Stop-hook 续期提供全新的单轮工具调用预算；使上限可配置 by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6238
* perf(glob): 在遍历期间剪枝忽略的目录，而非仅在事后过滤 by @minmax in https://github.com/QwenLM/qwen-code/pull/6123
* fix(web-shell): 为 @ 分组标题使用主题色 by @BZ-D in https://github.com/QwenLM/qwen-code/pull/6294
* fix(acp): 显式传递会话级设置，而非对 this.settings 竞态 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6292
* feat(web-shell): 从侧边栏管理会话（归档、取消归档、删除） by @wenshao in https://github.com/QwenLM/qwen-code/pull/6293
* fix(core): 改进 debug txt 诊断信息 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6277
* fix(ci): 要求维护者手动添加 `autofix/approved` 标签以启用 tier-1 快速路径 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6276
* fix(auth): 防止更改 API 密钥后持续出现 401 错误 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6284
* feat(daemon): 添加会话导出端点 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6297
* test(e2e): 使 Docker 沙箱可访问伪造的 OpenAI 服务 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6302
* fix(cli): 将长响应流式传输到滚动缓冲区，以停止滚动到顶部锁定 by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/6170
* fix(qqbot): 流式空闲刷新，增加工具调用和过期回调保护 by @Eric-GoodBoy-Tech in https://github.com/QwenLM/qwen-code/pull/6204
* fix(openai): 保留无描述的工具 by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6243
* feat(acp): 在初始化响应中声明 vision-bridge 图像能力 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6269
* fix(core): 对前台子代理强制执行代理并发上限 by @kagura-agent in https://github.com/QwenLM/qwen-code/pull/6300
* fix(ci): 停止对已关闭 PR 的审查机器人 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6304
* fix(core): 在 multiple_sentences 过滤器中跳过缩写 (#6077) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6193
* perf(cli): 基于 stat 失效机制按工作区缓存 LoadedSettings by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6310
* feat(web-shell): 支持紧凑版 echarts 完整数据块 by @zhangxy-zju in https://github.com/QwenLM/qwen-code/pull/6232
* fix(ci): 跳过过期的 PR 审查运行 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6313
* feat(acp): 批量会话加载重放 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6309
* fix(core): 将请求超时时间 0 视为禁用，而非立即中止 by @Nas01010101 in https://github.com/QwenLM/qwen-code/pull/6288
* fix(serve): 解决通过设置配置 API key 时预检请求中的虚假认证警告 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6296
* fix(core): 将 @-附加文件视为已读取，以执行前置读取检查 by @Nas01010101 in https://github.com/QwenLM/qwen-code/pull/6295
* fix(cli): 保留部分远程输入 JSONL 记录 by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6317
* fix(core): 在非 DashScope 服务器上通过 chat_template_kwargs 禁用 qwen 思考 by @beantownbytes in https://github.com/QwenLM/qwen-code/pull/6271
* feat(web-shell): 添加自定义 @提及面板 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6242
* feat(acp-bridge): 添加 EventBus 订阅者字节上限 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6314
* perf(ci): 优化 autofix 流水线 — 快速通道、跳过重复构建、作用域测试 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6315
* fix(web-shell): 启动新会话后保留技能斜杠命令 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6319
* test(core): 覆盖 truncated @-attachments 下 recordAttachedFileRead 的 full:false 分支 by @Nas01010101 in https://github.com/QwenLM/qwen-code/pull/6324
* feat(web-shell): 在 Daemon 状态页添加时间序列指标图表 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6307
* fix(web-shell): 对斜杠菜单中的内置命令和技能描述进行本地化 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6326
* fix(desktop): 强制执行 transform_data 隔离 by @gauravyad86 in https://github.com/QwenLM/qwen-code/pull/6285
* 扩展能力变更时通知模型 by @ZijianZhang989 in https://github.com/QwenLM/qwen-code/pull/6245
* fix(core): 跳过无操作的 max_tokens 升级 by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6234
* fix(core): 避免 OpenAPI schema 类型出现 null by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6323
* refactor(core): 集中化扩展运行时刷新 by @ZijianZhang989 in https://github.com/QwenLM/qwen-code/pull/6152
* feat(daemon): 添加会话组织功能 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6305
* feat(cli): 暴露 daemon 提示词队列状态 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6325
* 重启卡住的通道 ACP 桥接 by @qqqys in https://github.com/QwenLM/qwen-code/pull/6330
* docs(design): daemon 侧信道协调 (A1/A2/A4/A5) by @chiga0 in https://github.com/QwenLM/qwen-code/pull/4511
* docs: 修复技能调用语法并将飞书纳入通道列表 by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6320
* feat(scheduler): 可选按工具调用执行超时 by @minmax in https://github.com/QwenLM/qwen-code/pull/6124
* feat(web-shell): 添加 onSessionChange 和 onSubmitBefore 回调 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6333
* feat(core): 稳定工具 schema 声明顺序 by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/6339
* 修复自定义模型上下文窗口计算错误 by @heyparth1 in https://github.com/QwenLM/qwen-code/pull/6266
* feat(core): 模型回退链 — 过载时自动切换至备用模型 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6273
* fix(core): 将 OpenAI 推理过程保留为原始思考 by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6192
* fix(core): 在 Windows 上为 cmd.exe 添加 UTF-8 前缀 by @Aleks-0 in https://github.com/QwenLM/qwen-code/pull/6216
* feat(cli): 在文件系统边界检查中支持多文件夹工作区 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6278
* feat(web-shell): 支持提及标签的图标芯片 by @BZ-D in https://github.com/QwenLM/qwen-code/pull/6337
* 功能：LSP 服务器支持热重载 by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/5953
* fix(cli): 压缩期间允许排队输入 by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6336
* fix(web-shell): 完成延迟门控提交 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6342
* feat(cli): 添加大型管道帧测量 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6335
[codex] 添加主动通道循环工具 by @qqqys in https://github.com/QwenLM/qwen-code/pull/6287
* fix(cli): 更流畅的实时流预览 — 移除“正在生成更多”提示，暂存部分表格行 by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/6340
* perf: 记忆化技能扫描、防抖睡眠抑制器日志、保护 IDE readdir by @kagura-agent in https://github.com/QwenLM/qwen-code/pull/6155
* docs: 修复 settings.json 参考与 schema 的漂移问题 by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6351
* 修复(desktop)：保留已粘合的自动化历史记录 by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6344
* 新增(web-shell)：侧边栏支持命名会话组与颜色标签 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6350
* 修复(web-shell)：抑制过期的待处理提示符刷新错误 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6352
* 新增(web-shell)：新增定时任务管理页面 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6348
* 文档(web-shell)：补充图表渲染器集成说明 by @zhangxy-zju in https://github.com/QwenLM/qwen-code/pull/6353
* 新增(web-shell)：以内嵌面板形式展示设置与守护进程状态 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6341
* 持续集成(autofix)：将智能体提示词移至项目技能中 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6306
* 修复(web-shell)：限制虚拟滚动行数 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6362
* 修复(cli)：允许 ACP 本地回退读取 /tmp 目录 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6370
* 测试(core)：使上下文警告测试与默认令牌限制保持一致 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6391
* 修复(core)：默认上下文窗口调整为 200k by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6387
* 新增(web-shell)：在守护进程状态中新增令牌使用分析仪表板 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6388
* 新增(cli)：新增第一阶段工作区运行时注册表 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6394
* 修复(triage)：将测试文件排除在核心模块大小限制外，并区分新增与重构 by @yiliang114 with @Copilot in https://github.com/QwenLM/qwen-code/pull/6369
* 修复(cli)：在短终端中保持模型选择器条目连续 by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6359
* 修复(core)：在 OpenAI 错误日志中包含请求 ID by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6379
* 修复(cli)：在预提交 CI 中忽略当前审查运行 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6397
* 修复(autofix)：改进审查问题的处理与验证 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6382
* 修复(core)：要求 ReadFile 分页参数为整数 by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6381
* 新增(core)：将 PreToolUse 钩子的 'ask' 模式作为 TUI 确认提示暴露 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/5629
* 修复(core)：匹配条件规则和技能时解析符号链接 by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6371
* 修复(web-shell)：优化工具详情展示 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6399
* 修复(core)：保留流式传输空参数字符串且无参数的工具调用 by @tomsen02 in https://github.com/QwenLM/qwen-code/pull/6250
* 新增(review)：为 /review 命令新增问题保真度与根因归属检查门控 by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6395
* 修复(cli)：在通道代理中使用 EnvHttpProxyAgent 以尊重 NO_PROXY 设置 (#6401) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6405
* 修复(daemon)：在会话记录范围外处理设置重载事件 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6407
* 处理缺失的 web-shell 会话时不进行重定向 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6357
* 文档：补充 PreToolUse 钩子 permissionDecision "ask" 行为说明 by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6411
* 新增(cli)：新增第二阶段a工作区基础架构 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6410
* 修复(memory)：当智能体未执行任何工具调用时，不推进 AutoMemory 提取游标 by @lcheng321 in https://github.com/QwenLM/qwen-code/pull/6398
* 修复(core)：将图像负载替换限制在阈值内 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6380
* 新增(web-shell)：新增会话概览面板与窗口内分屏视图 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6400
* 修复(review)：从内置 /review 命令中移除 qwen-code 专属的核心基础设施门控 by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6412
* 修复(web-shell)：优化定时任务时间线 UI by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6386
* 修复(cli)：使流式表格渲染更流畅 by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/6345
* 修复(core)：限制大型 PDF 文本提取 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6409
* 新增(cli)：为 /model 命令新增 --project 和 --global 参数，支持按项目持久化模型配置 by @Alex-ai-future in https://github.com/QwenLM/qwen-code/pull/6060
* 性能优化(core)：新增会话启动性能分析器 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6349
* 新增(core)：新增 maxSubAgents 设置以限制并行子智能体数量 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6354
* 修复(core)：允许在压缩历史记录后回退 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6358
* 修复(core)：统一监控限制参数模式 by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6413
* 修复(core)：通过重新排序 reminderParts 防止 tool_search 时 KV 缓存失效 by @Aleks-0 in https://github.com/QwenLM/qwen-code/pull/6420
* 文档：将设计文档与计划统一整合至 docs/ 目录 by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6417
* 修复(shell)：在 Windows 上避免使用 Unix 分页器默认设置 by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6390
* 新增(scheduled-tasks)：使每个定时任务在独立的命名会话中运行 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6389
* 新增(cli)：支持堆叠式斜杠技能调用 by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6361
* 新增(core)：新增 Tool(param:value) 权限语法以实现参数级访问控制 by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6106
* fix(daemon): 保留用户消息来源元数据 by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6385
* feat(core): 新增 tools.visible 配置，用于在启动时选择性延迟显示工具 by @Aleks-0 in https://github.com/QwenLM/qwen-code/pull/6372
* fix(core): 防止重新调用已加载的技能时追加重复内容 by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6430
* fix(web-shell): 保持分屏视图会话列表的实时性，并在视图切换时保留面板 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6418
* feat(web-shell): 在侧边栏新建聊天按钮旁添加 Qwen 标志 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6437
* fix(web-shell): 优化用户标签和移动端菜单布局 by @yuanyuanAli in https://github.com/QwenLM/qwen-code/pull/6441
* fix(web-shell): 保持出错对话轮次处于展开状态 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6424
* fix(web-shell): 清除过期的浮动待办事项 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6425
* 避免在加载时刷新会话活动 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6439
* fix(web-shell): 在分屏面板状态中隐藏旋转加载提示语 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6447
* fix(core): 支持大文本范围读取 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6404
* fix(core): 从会话标题和回顾侧边查询提示中移除系统提醒块 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6435
* 升级 GitHub Actions 以兼容 Node 24 by @salmanmkc in https://github.com/QwenLM/qwen-code/pull/5157
* fix(autofix): 报告审查交接失败 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6415
* fix(monitor): 保留继承的 git 分页器 by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6429
* fix(serve): 对中断的模型流错误进行分类处理 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6422
* fix(web-shell): 优化工具调用摘要 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6450
* fix(web-shell): 修复分屏面板问题（移除“当前”徽章，发送时清空输入框） by @wenshao in https://github.com/QwenLM/qwen-code/pull/6454
* feat(web-shell): 为 Markdown 表格添加列重排、调整大小和冻结控制 by @jifeng in https://github.com/QwenLM/qwen-code/pull/6444
* feat(channels): 新增企业微信智能机器人渠道 by @qqqys in https://github.com/QwenLM/qwen-code/pull/6436
* [codex] 添加自然渠道记忆意图 by @qqqys in https://github.com/QwenLM/qwen-code/pull/6376
* perf(cli): 延迟启动预取任务 by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/6303
* feat(web-shell): 统一定时任务会话 — 绑定聊天创建的任务并添加时钟图标 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6453
* docs(channels): 在渠道概览中添加企业微信 by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6490
* feat(cli): 新增 serve 环境隔离与全局准入控制 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6416
* feat(cli): 支持通过内联预览、编辑器交接和对话内关闭开关来审查自动生成的技能 by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6393
* fix(core): 拒绝小数形式的 LSP 限制输入 by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6455
* fix(core): 将钩子显示名称匹配器与工具 ID 进行匹配 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6373
* fix(web-shell): 宽度不足时隐藏侧边栏设置文本 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6494
* fix(web-shell): 在守护进程状态使用仪表板中统计守护进程会话 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6493
* feat(cli): 在页脚为 DEFAULT 模式显示权限模式徽章 by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6498
* fix(channel): 转发 ACP 权限请求 by @qqqys in https://github.com/QwenLM/qwen-code/pull/6446
* fix(core): 拒绝 Windows 风格的工作区工件路径 by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6483
* fix(cli): 在单个可折叠工具的紧凑工具摘要中显示文件路径 by @Alex-ai-future in https://github.com/QwenLM/qwen-code/pull/6448
* fix(cua-driver): 迁移 Windows 脚本并重写 README by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6515
* feat(serve): 限制回放快照历史记录 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6482
* feat(web-shell): 在分屏面板中恢复完整输入框 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6510
* fix(cli): 限制实时流式表格的待处理高度（修复滚动到顶部锁定、卡住后转储、表头闪烁问题） by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/6421
* fix(cli): 在延迟超时后清理 IDE 客户端 by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/6509
* fix(web-shell): 防止侧边栏页脚溢出 by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6522
* fix(web-shell): 优化 Markdown 表格交互 by @jifeng in https://github.com/QwenLM/qwen-code/pull/6500
* fix(web-shell): 对 15 个文件中的约 43 个硬编码英文字符串进行国际化 (i18n) 处理 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6516
* fix(cli): 在会话模型上保持状态栏 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6514
* feat(hooks): 将后台任务和定时任务状态注入到 Stop/SubagentStop 钩子负载中 by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6531
* fix(cli): 允许在无 Bearer Token 的情况下更改审批模式 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6527
* fix(memory): 允许 forget 命令移除用户管理的记忆 by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6432
* feat(cli): 启用多工作区会话路由 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6511
* fix(core): 为 Claude 4.8+ 省略已弃用的 temperature 参数 by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6520
* feat(cli): 当服务端口被占用时自动重试下一个端口 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6513
* feat: 扩展文件重载 —— 监听插件更改并热重载运行时 by @ZijianZhang989 in https://github.com/QwenLM/qwen-code/pull/6347
* feat(channels): 添加 dmPolicy 配置以禁用私聊/DM 消息 by @chenghuichen in https://github.com/QwenLM/qwen-code/pull/6521
* feat(web-shell): 暴露外部拆分控件 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6523
* fix(core): 在 OpenAI 流式管道中检测非 SSE HTTP 200 响应 by @yiliang114 with @Copilot in https://github.com/QwenLM/qwen-code/pull/6466
* fix(scripts): 在发布版本控制中优雅处理缺失的 NPM dist-tags (#6476) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6481
* feat(core): 为 Agent 工具添加 working_dir，以将子代理固定到现有 worktree by @wenshao in https://github.com/QwenLM/qwen-code/pull/6456
* feat(autofix): 将审查循环扩展至所有 dev-bot PR，添加实时触发器 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6528
* fix(cli): 将一分钟以下的耗时设为固定宽度，以消除状态栏抖动 by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6533
* fix(memory): 为每个关联的 git worktree分配独立的自动记忆根目录 by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6462
* fix(cli): 任务取消后解除 /clear 命令阻塞并显示阻塞原因 by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6499
* fix(web-shell): 稳定拆分视图面板中的斜杠命令 i18n 显示 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6546
* 停止重复的子代理工具调用循环 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6543
* fix(session): 检测并标记断裂的历史记录链，而非静默截断 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6502
* fix(cli): 优先匹配命令名称而非别名，忽略 recentScore by @Alex-ai-future in https://github.com/QwenLM/qwen-code/pull/6504
* feat(memory): 使后台记忆代理的超时时间可配置 by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6459
* feat(cli): 为工作区运行时添加会话所有者索引 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6540
* docs: 修复模型提供商配置结构并刷新功能/设置漂移 by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6552
* fix(channels): 添加聊天负载诊断 by @qqqys in https://github.com/QwenLM/qwen-code/pull/6539
* fix(core): 可配置视觉桥接超时 + 使用新预算重试 by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6541
* fix(shell): 避免 pgrep 选择器导致自我终止 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6544
* 将浏览器自动化 MCP 限制在外部适配器上 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6472
* perf(core): 为文本 token 估算添加纯 ASCII 快速路径 by @dexhunter in https://github.com/QwenLM/qwen-code/pull/6551
* fix(extension): 在 Windows 上回退 git clone 前清理 tempDir by @C0d3N1nja97342 in https://github.com/QwenLM/qwen-code/pull/6545
* chore(core): 从 tool-search 测试中移除过时的 refreshStartupContextReminder 模拟 by @Aleks-0 in https://github.com/QwenLM/qwen-code/pull/6423
* fix(cli): 使记忆对话框与托管记忆保持一致 by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6434
* ci(autofix): 添加单目标调度器 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6547
* fix(daemon): 暴露工作区记忆任务错误详情 by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6431
* fix(serve): 停止 cdp-mcp-command 直接读取 process.env by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/6562
* fix(mobile-mcp): 从 UI 层次结构转储中移除 bounds by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6568
* fix(web-shell): 使对话框遮罩 z-index 可配置 by @cxruan in https://github.com/QwenLM/qwen-code/pull/6572
* feat(web-shell): 优化统计表格布局与待办面板 UI by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6559
* feat(cli): 列出可信工作区的持久化会话 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6558
* 为 web-shell 包添加工具基础设施 by @harbin1053020115 in https://github.com/QwenLM/qwen-code/pull/6517
* 修复禁用扩展和 ACP 预热的工作区技能 by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6534
* fix(ci): 为 VSCode IDE Companion 发布步骤添加重试逻辑 by @yiliang114 with @Copilot in https://github.com/QwenLM/qwen-code/pull/6574
* feat(core): 当文本提取溢出或失败时，将 PDF 页面渲染为图片 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6585
* fix(ci): 通过空响应检查检测静默分拣失败 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6566
* 修复长会话时间线滚动问题 by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6526
* feat(scheduled-tasks): 通过 create_sub_session 工具添加隔离运行模式 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6535
* feat(cli): VP 模式 —— 点击展开内联思考 + 自动隐藏滚动条 by @chiga0 in https://github.com/QwenLM/qwen-code/pull/6079
* feat(review): 将 Suggestion 发现结果发布为内联评论 by @wenshao in https://github.com/QwenLM/qwen-code/pull/6593
* fix(cli): 将用户输入转发至未声明参数的 MCP 提示词 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6571
* feat(daemon): 在重启间持久化会话工件 by @chiga0 in https://github.com/QwenLM/qwen-code/pull/6557
* feat(cli): 为 serve --channel 添加通道工作器设置重载 by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6598
* 支持 ACP 音频提示的 voiceBridge，由 @qqqys 在 https://github.com/QwenLM/qwen-code/pull/6576 中贡献
* ci(autofix): 实现按问题并发、路由取消进行中的任务及分配触发器，由 @yiliang114 在 https://github.com/QwenLM/qwen-code/pull/6609 中贡献
* fix(cua-driver): 完善缩放/滚动/鼠标工具的坐标归一化，由 @LaZzyMan 在 https://github.com/QwenLM/qwen-code/pull/6610 中贡献
* chore(cua-driver): 将版本引用更新为 0.7.1 + 添加修复文档，由 @LaZzyMan 在 https://github.com/QwenLM/qwen-code/pull/6616 中贡献
## 新贡献者
* @barry166 在 https://github.com/QwenLM/qwen-code/pull/6274 中首次贡献
* @minmax 在 https://github.com/QwenLM/qwen-code/pull/6123 中首次贡献
* @Nas01010101 在 https://github.com/QwenLM/qwen-code/pull/6288 中首次贡献
* @beantownbytes 在 https://github.com/QwenLM/qwen-code/pull/6271 中首次贡献
* @heyparth1 在 https://github.com/QwenLM/qwen-code/pull/6266 中首次贡献
* @Aleks-0 在 https://github.com/QwenLM/qwen-code/pull/6216 中首次贡献
* @guiyang882 在 https://github.com/QwenLM/qwen-code/pull/6367 中首次贡献
* @tomsen02 在 https://github.com/QwenLM/qwen-code/pull/6250 中首次贡献
* @chenghuichen 在 https://github.com/QwenLM/qwen-code/pull/6521 中首次贡献
* @zjunothing 在 https://github.com/QwenLM/qwen-code/pull/6533 中首次贡献
* @dexhunter 在 https://github.com/QwenLM/qwen-code/pull/6551 中首次贡献
* @C0d3N1nja97342 在 https://github.com/QwenLM/qwen-code/pull/6545 中首次贡献
* @cxruan 在 https://github.com/QwenLM/qwen-code/pull/6572 中首次贡献
* @harbin1053020115 在 https://github.com/QwenLM/qwen-code/pull/6517 中首次贡献
**完整更新日志**: https://github.com/QwenLM/qwen-code/compare/cua-driver-rs-v0.7.0...cua-driver-rs-v0.7.1
cua-driver prebuilt binaries — relative-coordinate fork
(vendored under `packages/cua-driver`).
- **macOS**: codesigned + notarized universal binary + `QwenCuaDriver.app`
- **Linux**: unsigned (x86_64 + arm64, glibc 2.31 floor)
- **Windows**: unsigned (x86_64 + arm64)
Enable relative coordinates: `CUA_DRIVER_RS_COORDINATE_SPACE=1`
(default `0` = off; optional `CUA_DRIVER_RS_COORDINATE_SCALE=1000`).
<!-- Release notes generated using configuration in .github/release.yml at cua-driver-rs-v0.7.1 -->
## What's Changed
* feat(scheduler): make recurring cron/loop job expiration configurable by @TianYuan1024 in https://github.com/QwenLM/qwen-code/pull/6173
* fix(serve): respect disabled skill settings by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6223
* fix: align vscode-ide-companion curly rule with root config by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6221
* fix(qqbot): security hardening — gateway validation, atomic state, sanitized logging by @Eric-GoodBoy-Tech in https://github.com/QwenLM/qwen-code/pull/6200
* fix(cua-driver): migrate Windows scripts + add /mcp Approve button by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6241
* [codex] Revert GLM tagged thinking parsing for DashScope by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6248
* fix(web-shell): improve session restore and loading feedback by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6220
* feat(mobile-mcp): vendor mobile-mcp with opt-in 0-1000 relative coordinates by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6235
* feat(web-shell): show the qwen-code version in the sidebar footer by @wenshao in https://github.com/QwenLM/qwen-code/pull/6222
* feat(daemon): add session artifact APIs by @chiga0 in https://github.com/QwenLM/qwen-code/pull/5895
* feat(web-shell): display nested sub-agents as a tree in the tasks panel by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6239
* Add sessionless workspace memory forget and dream by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6227
* fix: avoid vsce secret scanner false positive on regex patterns by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6247
* fix(web-shell): encode vision model picker selection & polish dispatch by @AmariahAK in https://github.com/QwenLM/qwen-code/pull/6236
* feat(web-shell): improve slash command discovery (taller menu, group counts, fuzzy search) by @wenshao in https://github.com/QwenLM/qwen-code/pull/6267
* fix(serve): Optimize daemon NDJSON stream handling by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6263
* ci(autofix): restore sandbox image flow by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6261
* fix(qqbot): markdown-first send, replyMsgId TTL, and dead code removal by @Eric-GoodBoy-Tech in https://github.com/QwenLM/qwen-code/pull/6201
* feat(daemon): expose visionModelId in workspace provider status and web-shell model dialog by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6262
* fix(triage): strengthen PR gate with batch detection, problem existence check, and red flag patterns by @pomelo-nwu in https://github.com/QwenLM/qwen-code/pull/5723
* feat(review): route suggestion-level findings to an updatable PR comment by @pomelo-nwu in https://github.com/QwenLM/qwen-code/pull/5786
* fix(autofix): unconditionally restore tracked files before branch checkout (#6281) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6286
* feat(serve): Add runtime.activity fields to daemon status API by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6270
* feat(web-shell): add a daemon status page backed by GET /daemon/status by @wenshao in https://github.com/QwenLM/qwen-code/pull/6272
* fix(vscode): keep auth quick inputs open on focus loss by @barry166 in https://github.com/QwenLM/qwen-code/pull/6274
* ci(audio): clarify macOS prebuild artifact suffix by @barry166 in https://github.com/QwenLM/qwen-code/pull/6275
* fix(cache): preserve tools prefix in side-query for Anthropic prompt-cache hits by @kagura-agent in https://github.com/QwenLM/qwen-code/pull/6225
* feat(web-shell): add MCP mentions and iconized @ references by @BZ-D in https://github.com/QwenLM/qwen-code/pull/6279
* fix(core): give Stop-hook continuations a fresh per-turn tool-call budget; make the cap configurable by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6238
* perf(glob): prune ignored directories during traversal, not just post-filter by @minmax in https://github.com/QwenLM/qwen-code/pull/6123
* fix(web-shell): use theme color for @ group titles by @BZ-D in https://github.com/QwenLM/qwen-code/pull/6294
* fix(acp): pass per-session settings explicitly instead of racing on this.settings by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6292
* feat(web-shell): manage sessions from the sidebar (archive, unarchive, delete) by @wenshao in https://github.com/QwenLM/qwen-code/pull/6293
* fix(core): improve debug txt diagnostics by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6277
* fix(ci): require maintainer-applied `autofix/approved` label for tier-1 fast-path by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6276
* fix(auth): prevent persistent 401 after API key change by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6284
* feat(daemon): Add session export endpoint by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6297
* test(e2e): make fake OpenAI reachable from Docker sandbox by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6302
* fix(cli): stream long responses into scrollback to stop scroll-to-top lock by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/6170
* fix(qqbot): streaming idle-flush with tool-call and stale-callback protection by @Eric-GoodBoy-Tech in https://github.com/QwenLM/qwen-code/pull/6204
* fix(openai): preserve descriptionless tools by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6243
* feat(acp): advertise vision-bridge image capability in initialize response by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6269
* fix(core): enforce agent concurrency cap on foreground sub-agents by @kagura-agent in https://github.com/QwenLM/qwen-code/pull/6300
* fix(ci): Stop review bots for closed PRs by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6304
* fix(core): skip abbreviations in multiple_sentences filter (#6077) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6193
* perf(cli): cache LoadedSettings per workspace with stat-based invalidation by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6310
* feat(web-shell): support compact echarts full data blocks by @zhangxy-zju in https://github.com/QwenLM/qwen-code/pull/6232
* fix(ci): skip stale PR review runs by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6313
* feat(acp): Batch session load replay by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6309
* fix(core): treat request timeout of 0 as disabled instead of aborting immediately by @Nas01010101 in https://github.com/QwenLM/qwen-code/pull/6288
* fix(serve): resolve false auth warning in preflight when API key is set via settings by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6296
* fix(core): treat @-attached files as read for prior-read enforcement by @Nas01010101 in https://github.com/QwenLM/qwen-code/pull/6295
* fix(cli): preserve partial remote input JSONL records by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6317
* fix(core): disable qwen thinking via chat_template_kwargs on non-DashScope servers by @beantownbytes in https://github.com/QwenLM/qwen-code/pull/6271
* feat(web-shell): add custom at mention panel by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6242
* feat(acp-bridge): Add EventBus subscriber byte cap by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6314
* perf(ci): optimize autofix pipeline — fast-track, skip duplicate build, scoped tests by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6315
* fix(web-shell): keep skill slash commands after starting a new session by @wenshao in https://github.com/QwenLM/qwen-code/pull/6319
* test(core): cover full:false branch of recordAttachedFileRead for truncated @-attachments by @Nas01010101 in https://github.com/QwenLM/qwen-code/pull/6324
* feat(web-shell): time-series metrics charts on Daemon Status by @wenshao in https://github.com/QwenLM/qwen-code/pull/6307
* fix(web-shell): localize built-in command and skill descriptions in the slash menu by @wenshao in https://github.com/QwenLM/qwen-code/pull/6326
* fix(desktop): enforce transform_data isolation by @gauravyad86 in https://github.com/QwenLM/qwen-code/pull/6285
* Notify model when extension capabilities change by @ZijianZhang989 in https://github.com/QwenLM/qwen-code/pull/6245
* fix(core): skip no-op max_tokens escalation by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6234
* fix(core): avoid null OpenAPI schema types by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6323
* refactor(core): centralize extension runtime refresh by @ZijianZhang989 in https://github.com/QwenLM/qwen-code/pull/6152
* feat(daemon): Add session organization by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6305
* feat(cli): Surface daemon prompt queue status by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6325
* Restart stalled ACP bridge for channels by @qqqys in https://github.com/QwenLM/qwen-code/pull/6330
* docs(design): daemon side-channel coordination (A1/A2/A4/A5) by @chiga0 in https://github.com/QwenLM/qwen-code/pull/4511
* docs: fix skill invocation syntax and include Feishu in channel lists by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6320
* feat(scheduler): opt-in per-tool-call execution timeout by @minmax in https://github.com/QwenLM/qwen-code/pull/6124
* feat(web-shell): add onSessionChange and onSubmitBefore callbacks by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6333
* feat(core): stabilize tool schema declaration order by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/6339
* Fix incorrect context window calculation for custom models by @heyparth1 in https://github.com/QwenLM/qwen-code/pull/6266
* feat(core): model fallback chain — auto-switch to backup models on overload by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6273
* fix(core): preserve OpenAI reasoning as raw thoughts by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6192
* fix(core): add UTF-8 prefix for cmd.exe on Windows by @Aleks-0 in https://github.com/QwenLM/qwen-code/pull/6216
* feat(cli): support multi-folder workspaces in file system boundary checks by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6278
* feat(web-shell): support icon chips for mention tags by @BZ-D in https://github.com/QwenLM/qwen-code/pull/6337
* Feat: LSP Server support hot reload by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/5953
* fix(cli): allow queued input during compression by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6336
* fix(web-shell): finalize deferred gated submissions by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6342
* feat(cli): Add large pipe frame measurement by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6335
* [codex] add proactive channel loop tools by @qqqys in https://github.com/QwenLM/qwen-code/pull/6287
* fix(cli): smoother live streaming preview — drop "generating more" cue, hold back partial table rows by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/6340
* perf: memoize skill scans, debounce sleep-inhibitor log, guard IDE readdir by @kagura-agent in https://github.com/QwenLM/qwen-code/pull/6155
* docs: fix settings.json reference drift against schema by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6351
* fix(desktop): preserve glued automation history records by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6344
* feat(web-shell): named session groups and color tags in the sidebar by @wenshao in https://github.com/QwenLM/qwen-code/pull/6350
* fix(web-shell): suppress stale pending prompt refresh errors by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6352
* feat(web-shell): add a Scheduled Tasks management page by @wenshao in https://github.com/QwenLM/qwen-code/pull/6348
* docs(web-shell): document chart renderer integration by @zhangxy-zju in https://github.com/QwenLM/qwen-code/pull/6353
* feat(web-shell): show Settings and Daemon Status as an in-place panel by @wenshao in https://github.com/QwenLM/qwen-code/pull/6341
* ci(autofix): move agent prompts into a project skill by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6306
* fix(web-shell): constrain virtual scroll rows by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6362
* fix(cli): Allow ACP local fallback reads from /tmp by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6370
* test(core): keep context warning test aligned with default token limit by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6391
* fix(core): default context windows to 200k by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6387
* feat(web-shell): add token-usage analytics dashboard to Daemon Status by @wenshao in https://github.com/QwenLM/qwen-code/pull/6388
* feat(cli): Add Phase 1 workspace runtime registry by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6394
* fix(triage): exclude test files from core module size gate and distinguish feat from refactor by @yiliang114 with @Copilot in https://github.com/QwenLM/qwen-code/pull/6369
* fix(cli): Keep model picker entries contiguous in short terminals by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6359
* fix(core): Include request IDs in OpenAI error logs by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6379
* fix(cli): ignore current review run in presubmit CI by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6397
* fix(autofix): improve review addressing and verification by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6382
* fix(core): require integer ReadFile pagination params by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6381
* feat(core): surface PreToolUse hook 'ask' as a TUI confirmation by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/5629
* fix(core): resolve symlinks when matching conditional rules and skills by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6371
* fix(web-shell): refine tool detail presentation by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6399
* fix(core): preserve no-argument tool calls that stream an empty arguments string by @tomsen02 in https://github.com/QwenLM/qwen-code/pull/6250
* feat(review): add issue-fidelity and root-cause ownership gate to /review by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6395
* fix(cli): use EnvHttpProxyAgent in channel proxy to respect NO_PROXY (#6401) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6405
* fix(daemon): Handle settings reload events outside transcript by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6407
* Handle missing web-shell sessions without redirecting by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6357
* docs: document PreToolUse hook permissionDecision "ask" behavior by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6411
* feat(cli): Add Phase 2a workspace foundation by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6410
* fix(memory): don't advance AutoMemory extract cursor when the agent makes zero tool calls by @lcheng321 in https://github.com/QwenLM/qwen-code/pull/6398
* fix(core): gate image payload replacement behind threshold by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6380
* feat(web-shell): add Session Overview panel and in-window split view by @wenshao in https://github.com/QwenLM/qwen-code/pull/6400
* fix(review): remove qwen-code-specific core-infra gate from bundled /review by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6412
* fix(web-shell): polish scheduled task timeline UI by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6386
* fix(cli): smoother streaming table rendering by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/6345
* fix(core): Gate large PDF text extraction by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6409
* feat(cli): add --project and --global flags to /model for per-project model persistence by @Alex-ai-future in https://github.com/QwenLM/qwen-code/pull/6060
* perf(core): Add session start profiler by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6349
* feat(core): add maxSubAgents setting to limit parallel sub-agent count by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6354
* fix(core): allow rewind after compressed history by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6358
* fix(core): align monitor limit parameter schemas by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6413
* fix(core): prevent KV-cache invalidation on tool_search by reordering reminderParts by @Aleks-0 in https://github.com/QwenLM/qwen-code/pull/6420
* docs: consolidate design docs and plans under docs/ by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6417
* fix(shell): avoid Unix pager default on Windows by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6390
* feat(scheduled-tasks): run each task in its own dedicated, named session by @wenshao in https://github.com/QwenLM/qwen-code/pull/6389
* feat(cli): support stacked slash-skill invocations by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6361
* feat(core): add Tool(param:value) permission syntax for parameter-level access control by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6106
* fix(daemon): preserve user message source metadata by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6385
* feat(core): add tools.visible config for selective deferred-tool visibility at startup by @Aleks-0 in https://github.com/QwenLM/qwen-code/pull/6372
* fix(core): prevent re-invoking loaded skill from appending duplicate content by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6430
* fix(web-shell): keep split-view session list fresh and preserve panes across view switches by @wenshao in https://github.com/QwenLM/qwen-code/pull/6418
* feat(web-shell): add Qwen logo beside the sidebar new-chat button by @wenshao in https://github.com/QwenLM/qwen-code/pull/6437
* fix(web-shell): Improve user tags and mobile menu layout by @yuanyuanAli in https://github.com/QwenLM/qwen-code/pull/6441
* fix(web-shell): keep errored turns expanded by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6424
* fix(web-shell): clear stale floating todos by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6425
* Avoid refreshing session activity on load by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6439
* fix(web-shell): hide rotating loading phrase in split-view pane status by @wenshao in https://github.com/QwenLM/qwen-code/pull/6447
* fix(core): Support large text range reads by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6404
* fix(core): strip system-reminder blocks from session title and recap side-query prompts by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6435
* Upgrade GitHub Actions for Node 24 compatibility by @salmanmkc in https://github.com/QwenLM/qwen-code/pull/5157
* fix(autofix): report review handoff failures by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6415
* fix(monitor): preserve inherited git pager by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6429
* fix(serve): classify interrupted model stream errors by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6422
* fix(web-shell): refine tool call summaries by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6450
* fix(web-shell): split-view pane fixes (remove "current" badge, clear composer on send) by @wenshao in https://github.com/QwenLM/qwen-code/pull/6454
* feat(web-shell): add column reorder, resize, and freeze controls to markdown table by @jifeng in https://github.com/QwenLM/qwen-code/pull/6444
* feat(channels): add WeCom intelligent robot channel by @qqqys in https://github.com/QwenLM/qwen-code/pull/6436
* [codex] add natural channel memory intents by @qqqys in https://github.com/QwenLM/qwen-code/pull/6376
* perf(cli): defer startup prefetch tasks by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/6303
* feat(web-shell): unify scheduled task sessions — bind chat-created tasks + clock icon by @wenshao in https://github.com/QwenLM/qwen-code/pull/6453
* docs(channels): add WeCom to channels overview by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6490
* feat(cli): Add serve env isolation and total admission by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6416
* feat(cli): review auto-generated skills with an inline preview, editor handoff, and an in-dialog off switch by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/6393
* fix(core): reject fractional LSP limit inputs by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6455
* fix(core): Match hook display-name matchers to tool ids by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6373
* fix(web-shell): hide sidebar settings text when width is insufficient by @wenshao in https://github.com/QwenLM/qwen-code/pull/6494
* fix(web-shell): count daemon sessions in Daemon Status usage dashboard by @wenshao in https://github.com/QwenLM/qwen-code/pull/6493
* feat(cli): Show permission mode badge in footer for DEFAULT mode by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6498
* fix(channel): Relay ACP permission requests by @qqqys in https://github.com/QwenLM/qwen-code/pull/6446
* fix(core): reject Windows-style workspace artifact paths by @VectorPeak in https://github.com/QwenLM/qwen-code/pull/6483
* fix(cli): show file path in compact tool summary for single collapsible tools by @Alex-ai-future in https://github.com/QwenLM/qwen-code/pull/6448
* fix(cua-driver): migrate Windows scripts + README rewrite by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6515
* feat(serve): Bound replay snapshot history by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6482
* feat(web-shell): restore the full composer in split-view panes by @wenshao in https://github.com/QwenLM/qwen-code/pull/6510
* fix(cli): bound the live streaming-table pending height (fix scroll-to-top lock, stall-then-dump, header flash) by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/6421
* fix(cli): clean up IDE client after deferred timeout by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/6509
* fix(web-shell): prevent sidebar footer overflow by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6522
* fix(web-shell): refine markdown table interactions by @jifeng in https://github.com/QwenLM/qwen-code/pull/6500
* fix(web-shell): i18n for ~43 hardcoded English strings across 15 files by @wenshao in https://github.com/QwenLM/qwen-code/pull/6516
* fix(cli): keep status line on session model by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6514
* feat(hooks): inject background tasks and cron jobs status into Stop/SubagentStop hook payloads by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/6531
* fix(cli): allow approval-mode changes without bearer token by @wenshao in https://github.com/QwenLM/qwen-code/pull/6527
* fix(memory): allow forget to remove user managed memory by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6432
* feat(cli): Enable multi-workspace session routing by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6511
* fix(core): omit deprecated temperature param for Claude 4.8+ by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6520
* feat(cli): auto-retry next port when serve port is in use by @wenshao in https://github.com/QwenLM/qwen-code/pull/6513
* feat: extension file reload — watch for plugin changes and hot-reload runtime by @ZijianZhang989 in https://github.com/QwenLM/qwen-code/pull/6347
* feat(channels): add dmPolicy config to disable private/DM messages by @chenghuichen in https://github.com/QwenLM/qwen-code/pull/6521
* feat(web-shell): expose external split controls by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6523
* fix(core): detect non-SSE HTTP 200 responses in OpenAI streaming pipeline by @yiliang114 with @Copilot in https://github.com/QwenLM/qwen-code/pull/6466
* fix(scripts): handle missing NPM dist-tags gracefully in release versioning (#6476) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6481
* feat(core): add working_dir to the Agent tool for pinning subagents to an existing worktree by @wenshao in https://github.com/QwenLM/qwen-code/pull/6456
* feat(autofix): extend review loop to all dev-bot PRs, add real-time triggers by @wenshao in https://github.com/QwenLM/qwen-code/pull/6528
* fix(cli): fixed-width elapsed time below one minute to stop status-line jitter by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6533
* fix(memory): give each linked git worktree its own auto-memory root by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6462
* fix(cli): unblock /clear after task cancellation and surface the blocked reason by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6499
* fix(web-shell): stabilize slash command i18n in split-view panes by @wenshao in https://github.com/QwenLM/qwen-code/pull/6546
* Stop repeated subagent tool-call loops by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6543
* fix(session): detect and mark broken history chains instead of silently truncating by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6502
* fix(cli): prefer command name match over alias match regardless of recentScore by @Alex-ai-future in https://github.com/QwenLM/qwen-code/pull/6504
* feat(memory): make background memory agent timeouts configurable by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6459
* feat(cli): Add session owner index for workspace runtimes by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6540
* docs: fix model-provider config shape and refresh feature/setting drift by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/6552
* fix(channels): add chat payload diagnostics by @qqqys in https://github.com/QwenLM/qwen-code/pull/6539
* fix(core): configurable vision bridge timeout + retry with fresh budget by @zjunothing in https://github.com/QwenLM/qwen-code/pull/6541
* fix(shell): avoid self-kill from pgrep selectors by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6544
* Gate browser automation MCP on external adapter by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6472
* perf(core): add pure-ASCII fast path to text token estimation by @dexhunter in https://github.com/QwenLM/qwen-code/pull/6551
* fix(extension): clean tempDir before fallback git clone on Windows by @C0d3N1nja97342 in https://github.com/QwenLM/qwen-code/pull/6545
* chore(core): remove stale refreshStartupContextReminder mocks from tool-search tests by @Aleks-0 in https://github.com/QwenLM/qwen-code/pull/6423
* fix(cli): align memory dialog with managed memory by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6434
* ci(autofix): Add single-target scheduler by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6547
* fix(daemon): surface workspace memory task error details by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6431
* fix(serve): stop cdp-mcp-command reading process.env directly by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/6562
* fix(mobile-mcp): strip bounds from UI hierarchy dump by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6568
* fix(web-shell): make dialog backdrop z-index configurable by @cxruan in https://github.com/QwenLM/qwen-code/pull/6572
* feat(web-shell): polish stats table layout and todo panel UI by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6559
* feat(cli): List persisted sessions for trusted workspaces by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6558
* Add harness infrastructure for web-shell package by @harbin1053020115 in https://github.com/QwenLM/qwen-code/pull/6517
* Fix workspace skills for disabled extensions and ACP preheat by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6534
* fix(ci): add retry logic to VSCode IDE Companion publish steps by @yiliang114 with @Copilot in https://github.com/QwenLM/qwen-code/pull/6574
* feat(core): render PDF pages to images when text extraction overflows or fails by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6585
* fix(ci): detect silent triage failures with empty-response check by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6566
* Fix long session timeline scrolling by @callmeYe in https://github.com/QwenLM/qwen-code/pull/6526
* feat(scheduled-tasks): add isolated run mode via create_sub_session tool by @wenshao in https://github.com/QwenLM/qwen-code/pull/6535
* feat(cli): VP mode — inline thought expand on click + auto-hiding scrollbar by @chiga0 in https://github.com/QwenLM/qwen-code/pull/6079
* feat(review): post Suggestion findings as inline comments by @wenshao in https://github.com/QwenLM/qwen-code/pull/6593
* fix(cli): forward user input to MCP prompts with no declared arguments by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6571
* feat(daemon): persist session artifacts across restarts by @chiga0 in https://github.com/QwenLM/qwen-code/pull/6557
* feat(cli): Add channel worker settings reload for serve --channel by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6598
* Support voiceBridge for ACP audio prompts by @qqqys in https://github.com/QwenLM/qwen-code/pull/6576
* ci(autofix): per-issue concurrency, route cancel-in-progress, assigned trigger by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6609
* fix(cua-driver): complete coordinate normalization for zoom/scroll/mouse tools by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6610
* chore(cua-driver): update version refs to 0.7.1 + add fix doc by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/6616
## New Contributors
* @barry166 made their first contribution in https://github.com/QwenLM/qwen-code/pull/6274
* @minmax made their first contribution in https://github.com/QwenLM/qwen-code/pull/6123
* @Nas01010101 made their first contribution in https://github.com/QwenLM/qwen-code/pull/6288
* @beantownbytes made their first contribution in https://github.com/QwenLM/qwen-code/pull/6271
* @heyparth1 made their first contribution in https://github.com/QwenLM/qwen-code/pull/6266
* @Aleks-0 made their first contribution in https://github.com/QwenLM/qwen-code/pull/6216
* @guiyang882 made their first contribution in https://github.com/QwenLM/qwen-code/pull/6367
* @tomsen02 made their first contribution in https://github.com/QwenLM/qwen-code/pull/6250
* @chenghuichen made their first contribution in https://github.com/QwenLM/qwen-code/pull/6521
* @zjunothing made their first contribution in https://github.com/QwenLM/qwen-code/pull/6533
* @dexhunter made their first contribution in https://github.com/QwenLM/qwen-code/pull/6551
* @C0d3N1nja97342 made their first contribution in https://github.com/QwenLM/qwen-code/pull/6545
* @cxruan made their first contribution in https://github.com/QwenLM/qwen-code/pull/6572
* @harbin1053020115 made their first contribution in https://github.com/QwenLM/qwen-code/pull/6517
**Full Changelog**: https://github.com/QwenLM/qwen-code/compare/cua-driver-rs-v0.7.0...cua-driver-rs-v0.7.1