feat(pmflow-os): add execution scope board

docs/PMFLOW_PRODUCT_DELIVERY_STANDARD.md

@@ -766,14 +766,14 @@ commit gate 判定：
 
 当前已完成：
 
-**main-flow-d4.5-execution-run-workspace-redesign：开发执行 Run 工作台重构。**
+**main-flow-d4.6-execution-scope-board：开发范围总览与需求执行队列。**
 
 原因：
 
-- `/execution-bridge` 已从模块堆叠重构为 `Execution Run Workspace`：顶部 Run Header、阶段 rail、左侧执行源包、中间当前阶段工作台、右侧 AI 决策面板、底部证据流水。
+- `/execution-bridge` 已补齐开发范围层：首屏先展示“本轮开发范围”和“需求执行队列”，再进入选中任务的 Run 工作台。
-- 执行页现在明确表达 `已确认需求 / 上下文与 Prompt 包 / 开发 / 测试 / 审查合并 / 回写` 的业务流转，所有主按钮由单点 Run 状态模型推导，避免散落按钮各自判断。
+- 页面现在按 Requirement 聚合正式 Task；同一需求下多个开发任务可在左侧切换，没有任务的已确认需求显示“尚未形成开发任务”，无法关联需求的任务进入“未归属任务”，不反推关系。
-- 手动回写从常驻表单收进 drawer；页面主界面优先展示阶段、源头、当前证据和下一步，不再像 dev-sync 调试录入台。
+- Run Header 现在同时展示范围规模与选中 Run 状态，右侧 AI 副驾增加范围缺口提示；主按钮仍由选中 Run 的单点状态模型推导。
-- 本轮只改前端信息架构和测试，复用既有 `execution-summary / execution-capabilities / execution-progress-events`，不新增后端合同，不执行命令，不自动验收、合并或写 Project Memory。
+- 本轮继续只改前端信息架构和测试，复用既有 `truth-state / execution-summary / execution-capabilities / execution-progress-events`，不新增后端合同，不执行命令，不自动验收、合并或写 Project Memory。
 
 ### 7.1 下一步推荐 Packet：`main-flow-d5-execution-acceptance-loop`
 
@@ -808,6 +808,47 @@ commit gate 判定：
 
 ## 8. 进度台账
 
+### 2026-05-06 / main-flow-d4.6-execution-scope-board / implemented
+
+完成内容：
+
+- 在 `/execution-bridge` 增加“本轮开发范围”总览，展示已确认需求、正式任务、执行中、待生成包、待测试/审查与阻塞数量。
+- 增加“需求执行队列”，按 Requirement 聚合开发任务；同一需求下多个 Task 可以切换到对应 Run 详情。
+- 对无法从 `state.requirements/state.tasks` 关联到需求的执行任务，进入“未归属任务”区，明确展示但不反推业务关系。
+- 保留 D4.5 的单 Run 工作台作为选中详情：中间展示当前任务的阶段 rail 与工作台，右侧展示执行源包、AI 阶段决策与范围缺口。
+- 新增前端只读范围模型，合并 `truth-state` 的 Requirement/Task 与 `execution-summary.activeDevTasks`，再为选中任务生成 scoped execution summary；不新增后端 API。
+- 焦点测试覆盖多任务开发范围、需求队列、任务切换和切换后的 Run 主动作。
+
+涉及文件：
+
+- `frontend/src/features/pmflow-os/pages/ExecutionBridgePage.tsx`
+- `frontend/src/features/pmflow-os/store.api.test.tsx`
+- `docs/PMFLOW_PRODUCT_DELIVERY_STANDARD.md`
+- `docs/architecture/CODE_REFERENCE_INDEX.md`
+- `docs/architecture/ARCHITECTURE_INVENTORY.md`
+- `docs/architecture/RUNTIME_GUARDRAILS.md`
+
+验证：
+
+- `cd frontend && npx tsc --noEmit --pretty false --incremental false` 通过。
+- `cd frontend && npm run test -- src/features/pmflow-os` 通过：35 passed。
+- `cd frontend && npm run build` 通过。
+- Playwright 浏览器检查 `http://localhost:5178/execution-bridge?demo=1` 通过：桌面与移动端均无页面横向溢出，首屏可见“本轮开发范围 / 需求执行队列”，点击“前置企业接入权限说明”后能切换到对应 Run；截图落点：
+  - `output/playwright/execution-scope-board-d46-desktop.png`
+  - `output/playwright/execution-scope-board-d46-desktop-selected.png`
+  - `output/playwright/execution-scope-board-d46-mobile.png`
+  - `output/playwright/execution-scope-board-d46-mobile-selected.png`
+- 浏览器控制台仍有既有 `GET /api/ai-readiness/project-launch` 401，属于现有 AI readiness 探针鉴权噪声，本 Packet 未改该链路。
+
+未完成或风险：
+
+- Requirement -> Task -> Execution Run 关联当前仍优先消费现有前端读模型和执行摘要；后续若后端提供稳定 run/list 合同，应将 scoped summary 推导下沉到后端。
+- 本轮不新增真实验收、技术审查、合并或 Project Memory 回写状态合同，这仍属于 `main-flow-d5-execution-acceptance-loop`。
+
+下一步建议：
+
+- 进入 `main-flow-d5-execution-acceptance-loop`：在 D4.6 的范围视图基础上，补真实验收、技术审查、合并确认和项目记忆回写状态，供 Space next action 与 Release/Growth 门禁稳定消费。
+
 ### 2026-05-06 / main-flow-d4.5-execution-run-workspace-redesign / implemented
 
 完成内容：

docs/architecture/ARCHITECTURE_INVENTORY.md

@@ -213,7 +213,7 @@
 - `frontend/src/features/pmflow-os/store.tsx` 当前是 PMFlow OS 共享 truth store：`?projectId=<id>` 时通过 React Query 调用 `frontend/src/api/pmflowOs.ts` 读取真实 `GET /pmflow-os/projects/{project_id}/truth-state` 与 `GET /pmflow-os/projects/{project_id}/execution-summary`，真实模式下项目内 `sendMessage` 会写 thread message 并触发后端真实 AI extraction；显式 `proposed_items` extraction、ChangeItem `accept/edit/defer`、ChangeSet commit、review decision、Execution Handoff create/consume、execution progress writeback、release validation run create、release feedback ingest、单条 iteration recommendation -> ChangeSet draft 与同组 iteration recommendation 批量 -> ChangeSet draft 均调用 `/pmflow-os` 后端并 refetch；只有显式 `?demo=1` 或 `?dataSource=demo` 时才启用本地 mock 演示状态机。真实读取/动作失败会进入错误/空态，不静默 fallback 成 mock 成功；AI extraction 失败显示“结构化失败/待补充”，不创建模板 ChangeSet。
 - `frontend/src/features/pmflow-os/store.tsx` 仍保留演示用项目内 AI 结构化闭环状态机：`command_idle / ai_extracting / extraction_ready / review_open / writeback_success`。`frontend/src/features/pmflow-os/pages/StructuredExtractionPage.tsx` 是本轮结构化提取页，也是结构化读取检查层：从指挥台进入时主动作进入待确认或继续补充；从 `ChangeReviewPage.tsx` 进入时通过 `origin=change_review / returnTo=change_review / changeSetId / itemId` 保留审查上下文，主动作返回待确认队列，补充证据才回指挥台；已 committed 的轮次显示为结构化写入快照并转向项目记忆或新变更链路。`ChangeReviewPage.tsx` 是决策与待确认中心，真实模式下顶部识别统计、待确认列表与右侧 AI 副驾评估都只消费后端 `changeSets / changeItems / evidence`，没有真实待审项时显示空态，禁止用 demo 审查项补位；确认写入 Project Memory 前，`pending_review / accepted / edited / deferred` ChangeItem 均可继续 accept/edit/defer，只有 `committed / skipped` 或关闭轮次不可再操作；`MemoryWritebackResultPage.tsx` 是项目记忆写入结果页，`WorkGraphPage.tsx` 是项目工作台关系总览与对象投影视图，`ExecutionBridgePage.tsx` 是执行桥上下文包、交接按钮、运行端领取确认、PR/测试与回写状态页；`ReleaseGrowthPage.tsx` 真实模式读取 `execution-summary.releaseGate`、`latestValidationRun` 与 `releaseSignals`，从 branch、commit、PR、test、failed evidence 展示 `可发布 / 阻塞 / 需要人工确认 / 缺少证据` 判断、阻塞原因、下一步动作、发布验证记录和发布后 feedback/revenue/ROI/compliance/recommendation 信号，缺少发布后信号时显示 empty/warning，不执行发布命令、不展示 mock 发布结果；页面可展示 recommendation quality grouping，把真实 recommendation 单条或同组批量生成下一轮待确认 ChangeSet draft，生成后仍进入 Change Review accept/edit/defer，不自动 commit、不写 Project Memory、不创建正式任务；`ChangeReviewPage.tsx` 会展示 `iteration_recommendation_group` evidence 的来源组、feedback 数、recommendation 数、风险、影响和审查优先级；同时读取 Account notification preference helper，用 `releaseGate` 偏好控制发布门禁强调提醒，用 `failureAlert` 偏好控制失败强调提醒，偏好关闭时只静默提醒表达强度，不隐藏 release evidence、阻塞原因、审计状态、validation run 或 releaseSignals；静态发布/增长原型只保留在显式 demo 模式。A3 已接入后端 truth-state，C1 已接入真实 message/extraction/review 状态机，C2 已接入真实 ChangeSet commit，Phase D 已接入真实 execution-summary/dev-sync，D.2 已接入 `POST /execution-handoffs` 创建 PromptPackage + handoff_created progress event，P1 已接入 `POST /execution-handoffs/{prompt_package_id}/consume` 创建 handoff_consumed progress event 并把 execution status 投影为 coding，Release Growth P1 已接入 execution evidence gate，Release Validation P1 已接入 `POST /release-validation-runs` 创建 `ReleaseValidationRun` 快照，Release Feedback P1 已接入 `POST /release-feedback-ingests` 写入 `FeedbackIngest` 并生成 `IterationRecommendation`，Iteration Loop P1 已接入 `POST /iteration-recommendations/{recommendation_id}/change-set-drafts` 把 recommendation 回流到 Change Review，Iteration Loop P2 已接入 `POST /iteration-recommendation-groups/{group_key}/change-set-drafts` 并在 `releaseSignals.recommendationGroups` 暴露分组/去重/风险/影响合同；真实 AI extraction 已接入独立 `change_review` protected scene、`pmflow_change_review_extraction_v1` 运行合同、`PROMPT_PMFLOW_CHANGE_REVIEW_EXTRACTION` Prompt 资产与后端 `llm_service.generate_structured_object`，成功时生成 open ChangeSet、pending_review ChangeItems 与 thread message evidence，失败时 fail-closed，不把 mock 当真实失败 fallback，也不复用 Project Launch draft prompt/function。`ProjectMemoryPage.tsx` 在真实模式下展示后端 `truth_nodes + semantic_facts` 的 `projectMemory` 投影，accepted/edited ChangeItem 只有 commit 成功后才进入 Project Memory。`frontend/src/features/pmflow-os/components/ProjectStructuredLoop.tsx` 提供 `StatusLabel / EvidenceSnippet / ObjectChangeCard / ProjectConversationPanel / StructuredExtractionPanel / ProjectAICopilotPanel` 等闭环组件，并复用契约化 `InlineNotice` 表达“未确认不得写入”的局部提示。
 - Execution Result Protocol P1 将 Execution Bridge 的执行结果回写收敛为 `execution_result_protocol_v1`：`backend/app/schemas/pmflow_os.py` 接收 `execution_status` 与 `blocker_summary`，`backend/app/services/pmflow_os_projection_service.py` 写入 `ProgressEvent.payload.protocol_version` 与 `execution_result` envelope，同时保留 `branch_name / commit_sha / pr_number / test_summary / workflow_state` flat 字段兼容现有投影；`frontend/src/api/pmflowOs.ts` 和 `ExecutionBridgePage.tsx` 支持手动记录 status/blocker。该链路只记录 Codex/IDE/Local Bridge/GitHub agent 已产生的结果，不执行本机命令、不推送、不发布、不自动创建正式任务；`execution-summary`、`releaseGate` 与 Team Feed 均从 `ProgressEvent` 只读消费。
-- Main Flow D4/D4.5 已把 Execution Capability Graph 接入并重构 `/execution-bridge` 为开发执行 Run 工作台：`frontend/src/features/pmflow-os/store.tsx` 读取并刷新真实 `GET /pmflow-os/projects/{project_id}/execution-capabilities`，`ExecutionBridgePage.tsx` 通过单点 Run 状态模型推导 `已确认需求 / 上下文包 / 开发 / 测试 / 审查合并 / 回写` 阶段、当前主动作、证据数和阻塞原因；页面结构为 Run Header、阶段 rail、执行源包、当前阶段工作台、AI 阶段决策面板和底部证据流水。handoff 主按钮受正式 Task、待确认范围、repo/prompt package 与 reviewer gate 约束，技术审查责任缺失时 store 会阻止真实 handoff，缺 repo 时只允许表达交接草案；手动回写在 drawer 中记录已发生的 branch/commit/PR/test/blocker/status。该层仍只做能力判定、交接合同和结果回写展示，不执行本机命令、不触发真实 agent、不自动验收、不自动合并、不自动发布。
+- Main Flow D4/D4.5/D4.6 已把 Execution Capability Graph 接入并重构 `/execution-bridge` 为开发范围总览 + 选中 Run 工作台：`frontend/src/features/pmflow-os/store.tsx` 读取并刷新真实 `GET /pmflow-os/projects/{project_id}/execution-capabilities`，`ExecutionBridgePage.tsx` 先消费 `truth-state.requirements/tasks` 与 `execution-summary.activeDevTasks` 形成前端只读开发范围模型，展示本轮需求/任务总览、按 Requirement 聚合的需求执行队列、未归属任务和选中任务 Run 详情；选中 Run 继续通过单点状态模型推导 `已确认需求 / 上下文包 / 开发 / 测试 / 审查合并 / 回写` 阶段、当前主动作、证据数和阻塞原因。handoff 主按钮受正式 Task、待确认范围、repo/prompt package 与 reviewer gate 约束，技术审查责任缺失时 store 会阻止真实 handoff，缺 repo 时只允许表达交接草案；手动回写在 drawer 中记录已发生的 branch/commit/PR/test/blocker/status。该层仍只做能力判定、交接合同和结果回写展示，不执行本机命令、不触发真实 agent、不自动验收、不自动合并、不自动发布。
 - Project Knowledge Hub 读模型由 `GET /pmflow-os/projects/{project_id}/project-memory/overview`、`items`、`items/{item_id}`、`assets`、`context-packs`、`writebacks` 承载，后端在 `backend/app/services/pmflow_os_projection_service.py` 从 `truth_nodes / tasks / semantic_facts / pmflow_evidence_links / pmflow_thread_messages / pmflow_change_sets / pmflow_change_items` 聚合首页雷达、全部记忆、项目资产、上下文包和写入版本记录。前端对应 `ProjectMemoryPage.tsx`、`ProjectMemoryCollectionPages.tsx` 与 `projectMemoryUtils.ts`：列表时间使用短格式，抽屉显示完整时间和北京时间；首页卡片与二级列表单条点击打开右侧详情抽屉，展示来源证据、影响对象和 AI 执行动作跳转；全部记忆支持类型、阶段、来源、状态、搜索筛选。该读模型只读，资产可自动索引，进入项目真相层仍必须经过 Change Review 或可信规则确认。
 - `frontend/src/features/account-settings/AccountSettingsPanel.tsx` 是 `/me` 设置页实现：桌面端为个人资料摘要 + 分区设置面板，移动端为分组列表 + `MobileBottomSheet` 编辑；当前覆盖个人资料、默认语言、AI 回答深度、证据展示偏好、待确认、执行回写、发布门禁和失败提醒通知偏好，并在通知偏好区展示页面内提醒消费说明、提醒强度和证据展示配置。`frontend/src/api/accountSettings.ts` 是前端合同层；有 token 时优先调用 `backend/app/api/account_settings_api/`，失败时回落浏览器本机暂存；P2 新增 `shouldShowAccountNotification` 与 `getAccountNotificationConsumption`，将四类通知偏好映射为本地 UI 的待确认、Execution Bridge、Release/Growth 和失败提醒显示/静默状态；P3 新增 `getReleaseGateNotificationPreference(settings, event)`，把 Release/Growth 的 `release_gate_* / release_validation_*` 事件映射到 `releaseGate` 偏好，把 `failed_event` 映射到 `failureAlert` 偏好；P5/P6 的 `getPostReleaseNotificationPreference(settings, event)` 把发布后 `feedback_received / revenue_signal / compliance_blocked` 映射到 `releaseGate` 偏好，把 `iteration_recommendation / next_change_review` 映射到 `pendingConfirmation` 偏好和 `change_review` surface，并返回 `spaceReadable / evidencePolicy / reminderPolicy` 供 Main Flow 与 Space 消费；P7 将通知偏好扩展为 `reminderIntensity=silent|normal|prominent` 与 `evidenceDisplay=compact|expanded`，并新增 `getAccountReminderConsumption(settings, event)` 覆盖 `change_review_high_risk / iteration_draft_ready / release_blocked / execution_stalled`，输出 `notificationTransport=in_app_only / evidencePolicy=preserve_business_evidence / auditPolicy=preserve_audit_state`；P8 新增 `getReviewDecisionNotificationPreference(settings, event)` 覆盖 `review_requested / review_approved / review_changes_requested / review_rejected / high_risk_approval_required`，继续使用 `pendingConfirmation` 偏好和 `change_review` surface，并声明 reviewerRouting、approval、evidence、audit state 均必须保留。该合同只控制页面内提醒表达，不代表邮件、桌面推送或系统通知中心已接入；偏好关闭也不得隐藏业务证据、reviewerRouting、approval、recommendation、ChangeSet、执行状态或审计状态。`backend/app/schemas/account_settings.py` 定义账号偏好合同，`backend/app/services/account_settings_service.py` 当前复用既有 `integrations` 表按 `tenant_id + user_id` 保存 `account_settings_preferences` 配置包，`backend/app/api/router.py` 已挂载该 API；未改用户表、OAuth/JWT/auth core 或 PMFlow OS truth store。
 - `frontend/src/features/pmflow-os/store.tsx` 当前是 PMFlow OS 共享 truth store：`?projectId=<id>` 时通过 React Query 调用 `frontend/src/api/pmflowOs.ts` 读取真实 `GET /pmflow-os/projects/{project_id}/truth-state` 与 `GET /pmflow-os/projects/{project_id}/execution-summary`，真实模式下项目内 `sendMessage` 会写 thread message 并触发后端真实 AI extraction；显式 `proposed_items` extraction、ChangeItem `accept/edit/defer`、ChangeSet commit、review decision、AI feedback、Execution Handoff create/consume、execution progress writeback、release validation run create、release feedback ingest 与 iteration recommendation -> ChangeSet draft 均调用 `/pmflow-os` 后端并 refetch；AI feedback 写入 `pmflow_ai_feedback_events`，在 `truth-state.aiFeedback` 恢复当前用户选择，并由后端按 `prompt_key` 生成待复核 `JudgeRun / OptimizerCandidate` 质量优化候选，用户改回赞同时挂起此前候选，Prompt Asset quality summary 会聚合 PMFlow 反馈与优化候选；只有显式 `?demo=1` 或 `?dataSource=demo` 时才启用本地 mock 演示状态机。真实读取/动作失败会进入错误/空态，不静默 fallback 成 mock 成功；AI extraction 失败显示“结构化失败/待补充”，不创建模板 ChangeSet。

docs/architecture/CODE_REFERENCE_INDEX.md

@@ -126,9 +126,9 @@
 - 个人空间首页 / 空间层入口前端产品域：参考 `frontend/src/features/personal-space/`、`frontend/src/api/space.ts` 与薄入口 `frontend/src/pages/PersonalSpacePage.tsx`；登录后路由入口为 `/space`，未登录由 `frontend/src/auth/RouteBoundary.tsx` 跳转 `/login?redirect=/space`；首屏以 `nextActionQueue` 展示“我的下一步”，并保留执行准备度、继续推进、待我确认、发布关注、发布后反馈关注、下一轮迭代草稿关注、执行中 / AI 正在推进、暂存想法、草案箱、缺项目归属补全与最近项目明细；红框三块在首页只作为工作雷达预览，暂存想法、草案箱、最近项目均最多露出少量高优先级条目，完整处理分别进入 `/space/ideas`、`/space/drafts` 与 `/space/projects?view=recent`。`frontend/src/features/personal-space/SpaceCollectionPages.tsx` 承载二级页：`/space/ideas` 支持暂存想法筛选、搜索、补充、归档与转 Project Launch 草案，`/space/drafts` 支持待确认草案筛选、搜索与回主流程确认；这些页面仍使用空间壳，不直接写项目真相层。Space 同时展示 AI readiness 客户视图卡片，管理员可从侧边栏或顶部状态进入 `/ai-config?tab=setup`。空间级 AI 管家在 `frontend/src/features/personal-space/components/SpaceAIConcierge.tsx`，入口只挂 `/space`，向现有 `/rag/ask` 注入个人空间上下文包并复用 QA session 历史能力；它只能围绕个人空间、跨项目下一步、待确认、执行中、草案、暂存想法和 AI 能力状态工作，真实写操作必须回主流程确认或跳转，不得声称已执行。全局 `Layout` 不再挂载旧 `FloatingQAChat` 悬浮球，普通后台页不显示无上下文聊天球，项目页继续使用项目级 `AI 副驾`。Space 左侧导航只承载稳定空间目的地，当前保留“空间首页”“暂存想法”“草案箱”“所有项目”和管理员“AI 使用设置”。点击项目主动作进入 `/pmflow-os?projectId=<id>`，待确认项进入 `/change-review?projectId=<id>&changeSetId=<id>`，review attention 待审项进入 `/change-review?projectId=<id>&changeSetId=<id>`，reviewer 缺失 warning 进入已有 `/team-layer?projectId=<id>`，执行中与执行准备度项进入 `/execution-bridge?projectId=<id>`，发布关注和发布后反馈关注项进入 `/release-growth?projectId=<id>`；执行失败和回写失败项进入 `/execution-bridge?projectId=<id>`；Space 聚合项目摘要、PMFlow OS `truth-state.reviewerRouting / truth-state.releases / execution-summary.releaseGate / execution-summary.releaseSignals / execution-summary.recentProgressEvents / execution-summary.recentDeliveries / execution-capabilities` 和真实 `GET /space/saved-ideas`，并兼容读取旧 AI PMO `waiting_manual / draft_pending_confirm` 工作项；`missing_project_scope` 仅在 Space 选择项目后复用 `POST /ai-pmo/work-items/{id}/decision`，不新增旁路接口，不直接写项目真相层。缺数据时显示明确空态或 warning，不虚构 reviewer、审查意见、待审 ChangeSet、执行能力、工具连接、验证、反馈、收入、ROI、合规、iteration recommendation 或 ChangeSet draft；暂存想法是唯一允许 Space 写入的 pre-project seed，保存原始输入、来源、状态、准备度和 draft 关联，生成草案只进入 Project Launch human review；除此之外不直接写 Project Memory、ChangeSet 状态、PMFlowApproval、execution event、ReleaseValidationRun、FeedbackIngest、RevenueEvent 或 ComplianceCheckRun；空间层共享左侧栏、顶部条与空间用量组件在 `frontend/src/features/personal-space/components/SpaceChrome.tsx`，`/project-launch` 桌面首次新建壳也复用该空间 Chrome
 - 新项目启动台 / 首次进入页前端产品域：参考 `frontend/src/features/project-launch/` 与薄入口 `frontend/src/pages/ProjectLaunchPage.tsx`；真实模式登录后路由入口为 `/project-launch`，默认通过 `frontend/src/api/pmflowLaunch.ts` 调用后端 taskized draft job/confirm API；AI 草案创建链路是 `POST /pmflow-launch/draft-jobs -> GET /pmflow-launch/draft-jobs/{job_id} -> POST /pmflow-launch/drafts/{draft_id}/confirm`，补充一句走 `POST /pmflow-launch/drafts/{draft_id}/revise` 或创建新的 `merge_supplement` job，旧 `POST /ai-steward/project-from-text` 只返回 410 迁移错误；只有显式 `?demo=1` 或 `?dataSource=demo` 才允许匿名并启用 `mockData.ts` 本地演示状态机，真实模式的操作流、生成过程、审查建议、附件引用和计数从当前输入、job 状态与后端返回草案派生，不再使用周报 demo 文案兜底；真实模式下 AI 未就绪时先渲染 `AISetupRequiredPanel`，按权限展示记录想法、查看空间状态或平台 AI 能力处理 CTA，不提供空项目创建入口，避免在缺少结构化上下文时提前创建项目壳。401/无 token 通过 `frontend/src/lib/apiError.ts` 归一为 `auth_required/session_expired`，通过 `frontend/src/lib/pendingIntent.ts` 保留输入；`input_needs_more_info` 422 归一为 `input_invalid` 并停在补充输入失败态，用户可见失败态不得展示 `Request failed with status code 401` 或把 401 归为网络异常
 - 前端身份边界与产品化错误：参考 `frontend/src/auth/RouteBoundary.tsx`、`frontend/src/lib/rbac.ts`、`frontend/src/lib/apiError.ts`、`frontend/src/lib/pendingIntent.ts`；`/space`、真实 `/project-launch`、`/me`、主链 standalone、空间管理入口与平台运营入口均在前端边界收口。空间管理入口使用 workspace role，`/platform/*` 和平台治理入口使用 `platform_context.capabilities`，后端 JWT/RBAC 仍是最终校验；pending intent 只能恢复 project launch 输入或承接已经生成好的 Project Launch draft，不允许登录后自动重放真实写请求
-- PMFlow OS 共享 truth store 与主链状态动作：参考 `frontend/src/features/pmflow-os/store.tsx`；当前支持 `?projectId=<id>` 走 React Query 读取真实 `truth-state`、`execution-summary` 与 `execution-capabilities`，真实模式下 `sendMessage` 会先写 thread message，再调用 `createPMFlowExtraction` 触发后端真实 AI extraction，成功后 refetch 出 open ChangeSet / pending_review ChangeItems；显式 `proposed_items` extraction 仍保留给受控 draft 场景；ChangeItem `accept/edit/defer`、ChangeSet commit、review decision、AI feedback、Execution Handoff create/consume、execution progress writeback、release validation run create、release feedback ingest、单条 iteration recommendation -> ChangeSet draft 与同组 iteration recommendation 批量 -> ChangeSet draft 均调后端并 refetch，不本地 promote 入 Project Memory、不伪造 approval、execution event、发布验证记录、发布后信号或下一轮 draft；D4 后 `createExecutionHandoff` 除正式 Task gate 外会读取 `execution-capabilities`，在技术审查责任缺失时阻止 handoff，避免把 agent 执行绕过 reviewer gate；D4.5 后 `/execution-bridge` 的主按钮由页面级 Run 状态模型消费 store 数据统一推导，不再散落在上下文包、回写视图和 AI 副驾卡片里；AI feedback 通过 `createPMFlowAIFeedback` 写入 `pmflow_ai_feedback_events` 并在 `truth-state.aiFeedback` 恢复当前用户选择；AI extraction 失败在 CommandCenter / StructuredExtraction 显示“结构化失败/待补充”，不展示模板假候选；`?demo=1` 才启用本地 mock 演示状态机和本地动作；项目内 AI 结构化闭环状态机为 `command_idle / ai_extracting / extraction_ready / review_open / writeback_success`
+- PMFlow OS 共享 truth store 与主链状态动作：参考 `frontend/src/features/pmflow-os/store.tsx`；当前支持 `?projectId=<id>` 走 React Query 读取真实 `truth-state`、`execution-summary` 与 `execution-capabilities`，真实模式下 `sendMessage` 会先写 thread message，再调用 `createPMFlowExtraction` 触发后端真实 AI extraction，成功后 refetch 出 open ChangeSet / pending_review ChangeItems；显式 `proposed_items` extraction 仍保留给受控 draft 场景；ChangeItem `accept/edit/defer`、ChangeSet commit、review decision、AI feedback、Execution Handoff create/consume、execution progress writeback、release validation run create、release feedback ingest、单条 iteration recommendation -> ChangeSet draft 与同组 iteration recommendation 批量 -> ChangeSet draft 均调后端并 refetch，不本地 promote 入 Project Memory、不伪造 approval、execution event、发布验证记录、发布后信号或下一轮 draft；D4 后 `createExecutionHandoff` 除正式 Task gate 外会读取 `execution-capabilities`，在技术审查责任缺失时阻止 handoff，避免把 agent 执行绕过 reviewer gate；D4.5 后 `/execution-bridge` 的主按钮由页面级 Run 状态模型消费 store 数据统一推导，不再散落在上下文包、回写视图和 AI 副驾卡片里；D4.6 后 `/execution-bridge` 额外消费 `truth-state.requirements/tasks` 与 `execution-summary.activeDevTasks` 形成前端只读开发范围模型，按 Requirement 聚合任务、展示未归属任务并支持切换选中 Run；AI feedback 通过 `createPMFlowAIFeedback` 写入 `pmflow_ai_feedback_events` 并在 `truth-state.aiFeedback` 恢复当前用户选择；AI extraction 失败在 CommandCenter / StructuredExtraction 显示“结构化失败/待补充”，不展示模板假候选；`?demo=1` 才启用本地 mock 演示状态机和本地动作；项目内 AI 结构化闭环状态机为 `command_idle / ai_extracting / extraction_ready / review_open / writeback_success`
 - Account Settings / `/me` 前端产品域与 P1-P8 偏好合同：参考 `frontend/src/features/account-settings/AccountSettingsPanel.tsx`、`frontend/src/features/pmflow-os/pages/AccountPage.tsx`、`frontend/src/api/accountSettings.ts`、`backend/app/api/account_settings_api/`、`backend/app/schemas/account_settings.py` 与 `backend/app/services/account_settings_service.py`；当前覆盖个人资料、默认语言、AI 回答深度、证据展示偏好、待确认、执行回写、发布门禁和失败提醒通知偏好，移动端使用分组列表与 bottom sheet，账户退出仍复用 `AccountControls.tsx`，不改 OAuth/JWT/auth core；P1 后端合同提供 `GET /account-settings/me` 与 `PUT /account-settings/me`，已由 `backend/app/api/router.py` 挂入核心路由组，按当前 `tenant_id + user_id` 保存账号偏好，前端有 token 时优先账号同步，后端不可用时明确回落 `仅本机暂存 / 保存失败`；P2 在 `frontend/src/api/accountSettings.ts` 提供 `shouldShowAccountNotification` 与 `getAccountNotificationConsumption`，把通知偏好映射为页面内提醒消费模型；P3 新增 `getReleaseGateNotificationPreference(settings, event)` 供 Release/Growth 后续消费：`release_gate_*` 与 `release_validation_*` 使用 `releaseGate` 偏好，`failed_event` 使用 `failureAlert` 偏好；P5/P6 的 `getPostReleaseNotificationPreference(settings, event)` 供发布后反馈与下一轮待确认消费：`feedback_received / revenue_signal / compliance_blocked` 使用 `releaseGate` 偏好，`iteration_recommendation / next_change_review` 使用 `pendingConfirmation` 偏好并落到 `change_review` surface；P7 将每类通知偏好扩展为 `reminderIntensity=silent|normal|prominent` 与 `evidenceDisplay=compact|expanded`，并新增 `getAccountReminderConsumption(settings, event)` 覆盖 `change_review_high_risk / iteration_draft_ready / release_blocked / execution_stalled`，稳定返回 `notificationTransport=in_app_only / evidencePolicy=preserve_business_evidence / auditPolicy=preserve_audit_state`；P8 新增 `getReviewDecisionNotificationPreference(settings, event)` 覆盖 `review_requested / review_approved / review_changes_requested / review_rejected / high_risk_approval_required`，统一使用 `pendingConfirmation` 偏好和 `change_review` surface，返回 `reviewerRoutingPolicy=preserve_reviewer_routing / approvalPolicy=preserve_approval_state`，偏好关闭只静默页面内提醒，不隐藏 reviewerRouting、approval、evidence 或 audit state；ReleaseGrowthPage 当前消费 `releaseGate` 与 `failureAlert` 两个偏好，只静默强调型提醒，不隐藏 release evidence、阻塞原因、审计状态、validation run、releaseSignals 或下一轮 ChangeSet draft；这些 helper 只控制本地 UI 提醒消费合同，不代表真实邮件、桌面推送或系统通知中心已接入
-- PMFlow OS 产品页入口：参考 `frontend/src/features/pmflow-os/pages/CommandCenterPage.tsx`、`StructuredExtractionPage.tsx`、`ChangeReviewPage.tsx`、`MemoryWritebackResultPage.tsx`、`WorkGraphPage.tsx`、`ExecutionBridgePage.tsx`、`ReleaseGrowthPage.tsx`、`TeamLayerPage.tsx`、`ProjectMemoryPage.tsx`、`AccountPage.tsx`；`StructuredExtractionPage.tsx` 根据 `origin=change_review / returnTo=change_review / changeSetId / itemId` 区分指挥台读取检查、待确认证据检查和已写入快照，从待确认进入时主动作必须返回 `/change-review` 原轮次，不把用户强制带回指挥台；`ChangeReviewPage.tsx` 真实模式可对当前变更集请求审查、通过、要求修改或拒绝，展示审查状态、建议审查人、审查意见和高风险审批门禁，但不自动 commit，且在确认写入 Project Memory 前允许 `pending_review / accepted / edited / deferred` ChangeItem 继续 accept/edit/defer；`ExecutionBridgePage.tsx` 真实模式读取 `execution-summary` 与 `execution-capabilities`，D4.5 后页面形态为开发执行 Run 工作台：Run Header、`已确认需求 / 上下文包 / 开发 / 测试 / 审查合并 / 回写` 阶段 rail、左侧执行源包、中间当前阶段工作台、右侧 AI 阶段决策面板和底部证据流水；主动作由单点 Run 状态模型推导，必须受正式 Task、待确认范围、repo/prompt package 与 reviewer gate 约束，缺 repo 只能生成交接草案，技术审查责任缺失时不能发起 handoff；手动回写在 drawer 内记录已发生的 branch/commit/PR/test/blocker/status，不执行命令；执行结果只汇总运行端、PR/commit、测试、阻塞和下一步，不把测试通过自动视为验收完成；`TeamLayerPage.tsx` 真实模式读取 `state.reviewerRouting`、`state.feed` 与 `execution-capabilities`，以 `团队动态 / 决策台 / 角色收件箱 / 智能体工位 / AI 副驾` 展示团队协作态势、待审变更责任路由、执行能力与容量信号、个人执行准备度和工具准备度，只给责任建议、阻塞解释与跳转，不自动分派、不自动审查、不写 ChangeSet 或 Project Memory；决策台与待审路由卡片不得预留独立空白操作列，正文应使用完整卡片宽度，1-2 条路由按双列或单列密度展示；`ReleaseGrowthPage.tsx` 真实模式展示 `execution-summary.releaseGate`、execution evidence、阻塞原因、下一步动作、`latestValidationRun` 和 `releaseSignals`，缺证据或缺发布后信号时显示 empty/warning，不展示 mock 发布结果；页面可调用真实 API 创建 release validation run、摄取 release feedback，并展示真实 feedback、revenue、ROI、compliance 与 recommendation 风险态，从真实 recommendation 单条或同组批量生成下一轮待确认 ChangeSet draft；生成后仍进入 Change Review accept/edit/defer，不自动 commit、不写 Project Memory、不执行部署命令、不接外部支付、不接真实合规平台；所有真实证据和审计状态仍可见，静态高保真发布/增长数据仅保留在显式 demo 模式
+- PMFlow OS 产品页入口：参考 `frontend/src/features/pmflow-os/pages/CommandCenterPage.tsx`、`StructuredExtractionPage.tsx`、`ChangeReviewPage.tsx`、`MemoryWritebackResultPage.tsx`、`WorkGraphPage.tsx`、`ExecutionBridgePage.tsx`、`ReleaseGrowthPage.tsx`、`TeamLayerPage.tsx`、`ProjectMemoryPage.tsx`、`AccountPage.tsx`；`StructuredExtractionPage.tsx` 根据 `origin=change_review / returnTo=change_review / changeSetId / itemId` 区分指挥台读取检查、待确认证据检查和已写入快照，从待确认进入时主动作必须返回 `/change-review` 原轮次，不把用户强制带回指挥台；`ChangeReviewPage.tsx` 真实模式可对当前变更集请求审查、通过、要求修改或拒绝，展示审查状态、建议审查人、审查意见和高风险审批门禁，但不自动 commit，且在确认写入 Project Memory 前允许 `pending_review / accepted / edited / deferred` ChangeItem 继续 accept/edit/defer；`ExecutionBridgePage.tsx` 真实模式读取 `execution-summary` 与 `execution-capabilities`，D4.6 后页面形态为开发范围总览 + 需求执行队列 + 选中 Run 工作台：首屏展示本轮已确认需求、正式任务、执行中、待生成包、待测试/审查与阻塞数量，左侧按 Requirement 聚合任务、列出未归属任务并允许切换选中任务；选中任务详情继续展示 Run Header、`已确认需求 / 上下文包 / 开发 / 测试 / 审查合并 / 回写` 阶段 rail、当前阶段工作台、执行源包、AI 阶段决策面板和底部证据流水；主动作由单点 Run 状态模型推导，必须受正式 Task、待确认范围、repo/prompt package 与 reviewer gate 约束，缺 repo 只能生成交接草案，技术审查责任缺失时不能发起 handoff；手动回写在 drawer 内记录已发生的 branch/commit/PR/test/blocker/status，不执行命令；执行结果只汇总运行端、PR/commit、测试、阻塞和下一步，不把测试通过自动视为验收完成；`TeamLayerPage.tsx` 真实模式读取 `state.reviewerRouting`、`state.feed` 与 `execution-capabilities`，以 `团队动态 / 决策台 / 角色收件箱 / 智能体工位 / AI 副驾` 展示团队协作态势、待审变更责任路由、执行能力与容量信号、个人执行准备度和工具准备度，只给责任建议、阻塞解释与跳转，不自动分派、不自动审查、不写 ChangeSet 或 Project Memory；决策台与待审路由卡片不得预留独立空白操作列，正文应使用完整卡片宽度，1-2 条路由按双列或单列密度展示；`ReleaseGrowthPage.tsx` 真实模式展示 `execution-summary.releaseGate`、execution evidence、阻塞原因、下一步动作、`latestValidationRun` 和 `releaseSignals`，缺证据或缺发布后信号时显示 empty/warning，不展示 mock 发布结果；页面可调用真实 API 创建 release validation run、摄取 release feedback，并展示真实 feedback、revenue、ROI、compliance 与 recommendation 风险态，从真实 recommendation 单条或同组批量生成下一轮待确认 ChangeSet draft；生成后仍进入 Change Review accept/edit/defer，不自动 commit、不写 Project Memory、不执行部署命令、不接外部支付、不接真实合规平台；所有真实证据和审计状态仍可见，静态高保真发布/增长数据仅保留在显式 demo 模式
 - 团队层只读动态与协作态势：`TeamLayerPage.tsx` 直接消费 `ProjectTruthState.feed` 聚合的 review request、review decision、变更集状态、execution progress 与 release gate 事件，并结合 `reviewerRouting`、`execution-capabilities` 和容量计划生成首屏摘要；前端只做空态、摘要、中文化展示和跳转，不写审批、不写 ChangeSet、不写通知，也不在真实模式 fallback 到 mock 动态。
 - PMFlow OS 行为组件：参考 `frontend/src/features/pmflow-os/components/ProjectStructuredLoop.tsx`、`ProjectStageRail.tsx`、`ThreadComposer.tsx`、`ExtractionRail.tsx`、`ChangeGroupPanel.tsx`、`ObjectGraphSection.tsx`、`ContextPacketPanel.tsx`、`ExecutionTimeline.tsx`、`ReleaseGatePanel.tsx`、`TeamFeedPanel.tsx`、`ChangeReviewGroup.tsx`、`WorkbenchPanel.tsx`、`ExecutionBridgeCards.tsx`、`PMFlowOsFrame.tsx`、`AccountControls.tsx`、`FeedbackPrimitives.tsx`、`MobileBottomNav.tsx`；其中 `ProjectStructuredLoop.tsx` 承接 `StatusLabel / EvidenceSnippet / ObjectChangeCard / ProjectConversationPanel / StructuredExtractionPanel / ProjectAICopilotPanel`，`PMFlowOsFrame.tsx` 承接项目内面包屑和项目切换器，`AccountControls.tsx` 承接桌面头像菜单、退出登录确认与移动端 bottom sheet，`FeedbackPrimitives.tsx` 承接 `PageBanner / InlineNotice / HumanDecisionBar / ConfirmDialog / MobileBottomSheet`，`MobileBottomNav.tsx` 承接正式应用壳移动端固定底部导航
 - PMFlow OS 原型状态回归测试：参考 `frontend/src/features/pmflow-os/store.test.ts`；真实 API 动作回归参考 `frontend/src/features/pmflow-os/store.api.test.tsx`

docs/architecture/RUNTIME_GUARDRAILS.md

@@ -47,7 +47,7 @@
 - 技术层 fallback 仅允许用于重试、缓存、解析修复、传输兼容等基础设施环节，禁止上浮为用户可见业务结论。
 - AI Steward 与其它生成链的成功/失败判定必须优先使用结构校验、字段完整性和证据一致性。
 - PMFlow OS `ChangeReviewPage` 真实模式只能展示 `ProjectTruthState.changeSets / changeItems / evidence` 的后端投影；顶部统计、待确认列表和右侧 AI 副驾评估都必须由真实 ChangeItem/Evidence 派生。没有真实待审项时必须显示空态或待补充，禁止用 demo 的知识库搜索、混合召回、固定风险/建议/计数补位。`accepted / edited / deferred` 只表示候选项已被人工处理但尚未写入 Project Memory；只要 ChangeSet 未 committed 且 ChangeItem 未 committed/skipped，就必须允许继续 accept/edit/defer，真正锁定边界是 ChangeSet commit。
-- PMFlow OS 团队审查与执行边界必须保持单向主链：`TeamLayerPage` 只允许消费 `reviewerRouting / reviewDecision / approvalState / TeamFeed` 做责任路由、角色收件箱、团队动态、审查状态、阻塞解释和跳转，不得直接执行 ChangeSet commit、写 Project Memory 或成为第二套 Change Review；Execution Bridge handoff 只能针对 Project Memory commit 后已进入正式项目对象的 Task，禁止从 `pending_review` ChangeItem、候选 Task、reviewer suggestion 或 team routing item 直接创建正式执行分派。`ExecutionBridgePage` 的 Run 工作台只能把 `已确认需求 / 上下文包 / 开发 / 测试 / 审查合并 / 回写` 作为前端流转表达和证据 gate，不得因为 UI 阶段推进而自动验收、自动合并、自动写 Project Memory 或自动进入发布成功。缺 reviewer、repo/runtime、PromptPackage、测试命令、权限或回写证据时必须返回 blocked/warning，禁止冒充已开始开发或已分配成功。
+- PMFlow OS 团队审查与执行边界必须保持单向主链：`TeamLayerPage` 只允许消费 `reviewerRouting / reviewDecision / approvalState / TeamFeed` 做责任路由、角色收件箱、团队动态、审查状态、阻塞解释和跳转，不得直接执行 ChangeSet commit、写 Project Memory 或成为第二套 Change Review；Execution Bridge handoff 只能针对 Project Memory commit 后已进入正式项目对象的 Task，禁止从 `pending_review` ChangeItem、候选 Task、reviewer suggestion 或 team routing item 直接创建正式执行分派。`ExecutionBridgePage` 的开发范围总览和 Run 工作台只能把 Requirement/Task 范围、`已确认需求 / 上下文包 / 开发 / 测试 / 审查合并 / 回写` 阶段作为前端流转表达和证据 gate；未归属任务只能展示为未归属，不得反推需求关系；不得因为 UI 阶段推进而自动验收、自动合并、自动写 Project Memory 或自动进入发布成功。缺 reviewer、repo/runtime、PromptPackage、测试命令、权限或回写证据时必须返回 blocked/warning，禁止冒充已开始开发或已分配成功。
 
 ### 3.3 业务动作归因主链
 - 运行时业务动作归因必须优先走：

frontend/src/features/pmflow-os/pages/ExecutionBridgePage.tsx

@@ -1,4 +1,4 @@
-import { useState } from "react";
+import { useEffect, useMemo, useState } from "react";
 import {
   AlertTriangle,
   CheckCircle2,
@@ -44,9 +44,12 @@ import {
 } from "@/features/pmflow-os/components/ExecutionBridgeCards";
 import { usePMFlowOsStore } from "@/features/pmflow-os/store";
 import type {
+  Requirement as PMFlowRequirement,
   PMFlowExecutionCapabilityMode,
   PMFlowExecutionCapabilitySnapshot,
   PMFlowExecutionModeCapability,
+  ProjectTruthState,
+  Task as PMFlowProjectTask,
 } from "@/features/pmflow-os/types";
 import { cn } from "@/lib/utils";
 
@@ -144,6 +147,51 @@ interface ExecutionRunState {
   sourceItems: Array<{ label: string; value: string; tone?: "default" | "warning" | "success" }>;
 }
 
+type ExecutionScopeTone = "ready" | "running" | "waiting" | "blocked" | "empty";
+
+interface ExecutionScopeTaskItem {
+  id: string;
+  title: string;
+  summary: string;
+  status: string;
+  stage: string;
+  requirementId?: string | null;
+  requirementTitle?: string | null;
+  devTask?: PMFlowDevTask | null;
+  projectTask?: PMFlowProjectTask | null;
+  runStage: string;
+  nextAction: string;
+  evidenceCount: number;
+  tone: ExecutionScopeTone;
+}
+
+interface ExecutionScopeRequirementGroup {
+  id: string;
+  title: string;
+  summary: string;
+  status: string;
+  tasks: ExecutionScopeTaskItem[];
+  coverageLabel: string;
+  tone: ExecutionScopeTone;
+}
+
+interface ExecutionScopeBoard {
+  groups: ExecutionScopeRequirementGroup[];
+  unassignedTasks: ExecutionScopeTaskItem[];
+  allTasks: ExecutionScopeTaskItem[];
+  selectedTaskId: string | null;
+  totals: {
+    requirements: number;
+    tasks: number;
+    running: number;
+    waitingPackage: number;
+    waitingTest: number;
+    waitingReview: number;
+    blocked: number;
+  };
+  nextQueueAction: string;
+}
+
 function isFormalDevTask(task?: PMFlowDevTask | null) {
   const status = String(task?.status || "").toLowerCase();
   return Boolean(
@@ -294,9 +342,23 @@ export default function ExecutionBridgePage() {
   const isLoading = dataSourceMode === "api" && executionSummaryStatus === "loading";
   const error = dataSourceMode === "api" ? executionSummaryError : null;
   const [writebackOpen, setWritebackOpen] = useState(false);
-  const run = summary ? buildExecutionRunState(summary, capabilities) : null;
+  const [selectedScopeTaskId, setSelectedScopeTaskId] = useState<string | null>(null);
+  const baseScope = useMemo(() => (summary ? buildExecutionScopeBoard(state, summary, capabilities, selectedScopeTaskId) : null), [state, summary, capabilities, selectedScopeTaskId]);
+  const selectedTaskId = baseScope?.selectedTaskId || null;
+  const scopedSummary = useMemo(
+    () => (summary && selectedTaskId ? createScopedExecutionSummary(summary, state, selectedTaskId) : summary),
+    [summary, state, selectedTaskId]
+  );
+  const scope = useMemo(() => (scopedSummary ? buildExecutionScopeBoard(state, scopedSummary, capabilities, selectedTaskId) : baseScope), [state, scopedSummary, capabilities, selectedTaskId, baseScope]);
+  const run = scopedSummary ? buildExecutionRunState(scopedSummary, capabilities) : null;
+
+  useEffect(() => {
+    if (!scope?.selectedTaskId || selectedScopeTaskId) return;
+    setSelectedScopeTaskId(scope.selectedTaskId);
+  }, [scope?.selectedTaskId, selectedScopeTaskId]);
+
   const handlePrimaryAction = () => {
-    if (!summary || !run || !run.primaryAction.enabled) return;
+    if (!scopedSummary || !run || !run.primaryAction.enabled) return;
     if (run.primaryAction.kind === "create_handoff" && !isDemo && run.formalTask) {
       createExecutionHandoff({
         task_id: run.formalTask.id,
@@ -319,25 +381,26 @@ export default function ExecutionBridgePage() {
       return;
     }
     if (run.primaryAction.kind === "change_review") {
-      window.location.assign(`/change-review?projectId=${summary.projectId}`);
+      window.location.assign(`/change-review?projectId=${scopedSummary.projectId}`);
       return;
     }
     if (run.primaryAction.kind === "team_layer") {
-      window.location.assign(`/team-layer?projectId=${summary.projectId}`);
+      window.location.assign(`/team-layer?projectId=${scopedSummary.projectId}`);
       return;
     }
     if (run.primaryAction.kind === "release") {
-      window.location.assign(`/release-growth?projectId=${summary.projectId}`);
+      window.location.assign(`/release-growth?projectId=${scopedSummary.projectId}`);
     }
   };
 
   return (
     <div className="min-h-full bg-[#f7f9fd] pb-6 text-slate-950">
       <div className="mx-auto max-w-[1760px] space-y-4 px-4 py-4 lg:px-6">
-        {run && summary ? (
+        {run && scopedSummary ? (
           <ExecutionRunHeader
-            summary={summary}
+            summary={scopedSummary}
             run={run}
+            scope={scope}
             dataSourceMode={dataSourceMode}
             onRefresh={() => {
               refetchExecutionSummary();
@@ -372,38 +435,45 @@ export default function ExecutionBridgePage() {
           <ExecutionEmptyState title="缺少项目执行上下文" summary="URL 未提供 projectId，执行桥不会回退到假数据。请从真实项目进入或显式追加 ?demo=1。" />
         ) : null}
 
-        {summary && run ? (
+        {scopedSummary && run && scope ? (
           <>
-            <ExecutionRunStageRail run={run} />
             {dataSourceMode === "api" && executionCapabilitiesStatus === "loading" ? (
               <ExecutionEmptyState title="正在读取执行能力图谱" summary="主流程已根据执行摘要渲染，执行模式和责任门禁会在能力图谱同步后刷新。" />
             ) : null}
             {dataSourceMode === "api" && executionCapabilitiesStatus === "error" ? (
               <ExecutionEmptyState title="执行能力图谱同步失败" summary={executionCapabilitiesError || "执行能力信号暂未同步，主流程只展示执行摘要证据。"} />
             ) : null}
-            <ExecutionWarningList warnings={summary.warnings} />
+            <ExecutionWarningList warnings={scopedSummary.warnings} />
+            <ExecutionScopeOverview scope={scope} />
 
             <div className="grid gap-4 min-[1320px]:grid-cols-[360px_minmax(0,1fr)_340px]">
-              <ExecutionRunSourcePanel run={run} summary={summary} />
+              <ExecutionScopeQueuePanel scope={scope} selectedTaskId={selectedTaskId} onSelectTask={setSelectedScopeTaskId} />
-              <ExecutionRunWorkspacePanel
+              <div className="min-w-0 space-y-4">
-                run={run}
+                <ExecutionRunStageRail run={run} />
-                summary={summary}
+                <ExecutionRunWorkspacePanel
-                onOpenWriteback={() => setWritebackOpen(true)}
+                  run={run}
-                onDemoPush={isDemo ? advanceExecution : undefined}
+                  summary={scopedSummary}
-              />
+                  onOpenWriteback={() => setWritebackOpen(true)}
-              <ExecutionRunDecisionPanel
+                  onDemoPush={isDemo ? advanceExecution : undefined}
-                run={run}
+                />
-                summary={summary}
+              </div>
-                capabilities={capabilities}
+              <div className="space-y-4">
-                onPrimaryAction={handlePrimaryAction}
+                <ExecutionRunSourcePanel run={run} summary={scopedSummary} />
-                onOpenWriteback={() => setWritebackOpen(true)}
+                <ExecutionRunDecisionPanel
-              />
+                  run={run}
+                  summary={scopedSummary}
+                  scope={scope}
+                  capabilities={capabilities}
+                  onPrimaryAction={handlePrimaryAction}
+                  onOpenWriteback={() => setWritebackOpen(true)}
+                />
+              </div>
             </div>
 
-            <ExecutionProgressPanel summary={summary} />
+            <ExecutionProgressPanel summary={scopedSummary} />
             {writebackOpen && !isDemo ? (
               <ExecutionWritebackDrawer
-                summary={summary}
+                summary={scopedSummary}
                 onRecordProgress={recordExecutionProgress}
                 onClose={() => setWritebackOpen(false)}
               />
@@ -418,12 +488,14 @@ export default function ExecutionBridgePage() {
 function ExecutionRunHeader({
   summary,
   run,
+  scope,
   dataSourceMode,
   onRefresh,
   onPrimaryAction,
 }: {
   summary: PMFlowExecutionSummary;
   run: ExecutionRunState;
+  scope?: ExecutionScopeBoard | null;
   dataSourceMode: "api" | "demo" | "empty";
   onRefresh: () => void;
   onPrimaryAction: () => void;
@@ -444,6 +516,7 @@ function ExecutionRunHeader({
             {run.task?.title || "等待正式开发任务"} · 当前阶段：<span className="font-semibold text-slate-900">{run.currentStage.label}</span>
           </div>
           <div className="mt-4 flex flex-wrap items-center gap-2">
+            {scope ? <StatusChip label={`范围 ${scope.totals.requirements} 需求 / ${scope.totals.tasks} 任务`} tone="brand" /> : null}
             <StatusChip label={run.result.statusLabel} tone={run.result.tone} />
             <StatusChip label={run.mode?.label || "执行模式未同步"} tone={modeTone(run.mode?.status)} />
             <StatusChip label={run.formalTask ? "正式 Task" : "非正式任务"} tone={run.formalTask ? "success" : "warning"} />
@@ -505,13 +578,13 @@ function ExecutionRunStageRail({ run }: { run: ExecutionRunState }) {
         </div>
         <StatusChip label={run.currentStage.label} tone={stageToneToChip(run.currentStage.tone)} />
       </div>
-      <div className="mt-4 grid gap-3 lg:grid-cols-6">
+      <div className="mt-4 flex gap-3 overflow-x-auto pb-1">
         {run.stages.map((stage, index) => (
-          <div key={stage.key} className="relative">
+          <div key={stage.key} className="relative min-w-[150px] flex-1">
-            {index < run.stages.length - 1 ? <div className="absolute left-[calc(50%+18px)] top-5 hidden h-px w-[calc(100%-36px)] border-t border-dashed border-[#cfd8e7] lg:block" /> : null}
+            {index < run.stages.length - 1 ? <div className="absolute left-[calc(50%+18px)] top-5 hidden h-px w-[calc(100%-36px)] border-t border-dashed border-[#cfd8e7] min-[1700px]:block" /> : null}
             <div
               className={cn(
-                "relative rounded-lg border px-3 py-3",
+                "relative h-full rounded-lg border px-3 py-3",
                 stage.tone === "done" && "border-emerald-200 bg-emerald-50",
                 stage.tone === "current" && "border-[#aac4ff] bg-[#f3f7ff] shadow-[0_8px_20px_rgba(36,99,235,0.08)]",
                 stage.tone === "waiting" && "border-[#e8eef8] bg-white",
@@ -542,6 +615,129 @@ function ExecutionRunStageRail({ run }: { run: ExecutionRunState }) {
   );
 }
 
+function ExecutionScopeOverview({ scope }: { scope: ExecutionScopeBoard }) {
+  return (
+    <section className="rounded-xl border border-[#dbe6ff] bg-white px-5 py-4 shadow-[0_1px_2px_rgba(15,23,42,0.03)]">
+      <div className="flex flex-wrap items-start justify-between gap-4">
+        <div>
+          <div className="text-base font-semibold text-slate-900">本轮开发范围</div>
+          <p className="mt-1 text-sm leading-6 text-slate-500">
+            先看所有已确认需求和正式任务，再进入单个 Run；没有关联的任务会进入未归属队列，不反推关系。
+          </p>
+        </div>
+        <StatusChip label={scope.nextQueueAction} tone={scope.totals.blocked > 0 ? "warning" : "brand"} />
+      </div>
+      <div className="mt-4 grid gap-3 sm:grid-cols-2 lg:grid-cols-6">
+        <ExecutionMetric label="已确认需求" value={`${scope.totals.requirements}`} />
+        <ExecutionMetric label="正式任务" value={`${scope.totals.tasks}`} />
+        <ExecutionMetric label="执行中" value={`${scope.totals.running}`} tone={scope.totals.running ? "brand" : "default"} />
+        <ExecutionMetric label="待生成包" value={`${scope.totals.waitingPackage}`} tone={scope.totals.waitingPackage ? "warning" : "default"} />
+        <ExecutionMetric label="待测试/审查" value={`${scope.totals.waitingTest + scope.totals.waitingReview}`} tone={scope.totals.waitingTest + scope.totals.waitingReview ? "warning" : "default"} />
+        <ExecutionMetric label="阻塞" value={`${scope.totals.blocked}`} tone={scope.totals.blocked ? "warning" : "default"} />
+      </div>
+    </section>
+  );
+}
+
+function ExecutionScopeQueuePanel({
+  scope,
+  selectedTaskId,
+  onSelectTask,
+}: {
+  scope: ExecutionScopeBoard;
+  selectedTaskId: string | null;
+  onSelectTask: (taskId: string) => void;
+}) {
+  return (
+    <aside className="rounded-xl border border-[#e2e9f5] bg-white shadow-[0_1px_2px_rgba(15,23,42,0.03)]">
+      <div className="border-b border-[#edf2f8] px-5 py-4">
+        <div className="flex items-center gap-2">
+          <PackageCheck className="h-4 w-4 text-[#2463EB]" />
+          <h2 className="text-base font-semibold text-slate-900">需求执行队列</h2>
+        </div>
+        <p className="mt-2 text-sm leading-6 text-slate-500">按 Requirement 分组查看本轮开发范围，点击任务切换 Run 详情。</p>
+      </div>
+      <div className="max-h-[980px] space-y-4 overflow-auto p-5">
+        {scope.groups.length ? (
+          scope.groups.map((group) => (
+            <section key={group.id} className="rounded-lg border border-[#e8eef8] bg-[#fbfdff] p-3">
+              <div className="flex items-start justify-between gap-3">
+                <div className="min-w-0">
+                  <div className="truncate text-sm font-semibold text-slate-900">{group.title}</div>
+                  <p className="mt-1 line-clamp-2 text-xs leading-5 text-slate-500">{group.summary || "该需求暂无摘要。"}</p>
+                </div>
+                <StatusChip label={group.coverageLabel} tone={scopeToneToChip(group.tone)} />
+              </div>
+              <div className="mt-3 space-y-2">
+                {group.tasks.length ? (
+                  group.tasks.map((task) => (
+                    <ExecutionScopeTaskButton key={task.id} task={task} active={task.id === selectedTaskId} onClick={() => onSelectTask(task.id)} />
+                  ))
+                ) : (
+                  <div className="rounded-lg border border-dashed border-[#d8e3f4] bg-white px-3 py-3 text-xs leading-5 text-slate-500">
+                    尚未形成开发任务；需要先从 Project Memory 拆出正式 Task。
+                  </div>
+                )}
+              </div>
+            </section>
+          ))
+        ) : (
+          <ExecutionEmptyState title="暂无已确认需求" summary="当前项目记忆里没有可进入开发范围的已确认 Requirement。" />
+        )}
+
+        {scope.unassignedTasks.length ? (
+          <section className="rounded-lg border border-amber-200 bg-amber-50 p-3">
+            <div className="text-sm font-semibold text-amber-900">未归属任务</div>
+            <p className="mt-1 text-xs leading-5 text-amber-700">这些任务缺少 Requirement 关联，先展示但不反推需求关系。</p>
+            <div className="mt-3 space-y-2">
+              {scope.unassignedTasks.map((task) => (
+                <ExecutionScopeTaskButton key={task.id} task={task} active={task.id === selectedTaskId} onClick={() => onSelectTask(task.id)} />
+              ))}
+            </div>
+          </section>
+        ) : null}
+      </div>
+    </aside>
+  );
+}
+
+function ExecutionScopeTaskButton({
+  task,
+  active,
+  onClick,
+}: {
+  task: ExecutionScopeTaskItem;
+  active: boolean;
+  onClick: () => void;
+}) {
+  return (
+    <button
+      type="button"
+      className={cn(
+        "w-full rounded-lg border px-3 py-3 text-left transition",
+        active ? "border-[#9ebcff] bg-white shadow-[0_8px_18px_rgba(36,99,235,0.08)]" : "border-[#e8eef8] bg-white hover:border-[#bfd4ff]",
+        task.tone === "blocked" && "border-rose-200 bg-rose-50",
+        task.tone === "waiting" && !active && "bg-white",
+        task.tone === "running" && !active && "bg-[#fbfdff]"
+      )}
+      onClick={onClick}
+    >
+      <div className="flex items-start justify-between gap-3">
+        <div className="min-w-0">
+          <div className="truncate text-sm font-semibold text-slate-900">{task.title}</div>
+          <div className="mt-1 text-xs leading-5 text-slate-500">{task.runStage} · {task.nextAction}</div>
+        </div>
+        <StatusChip label={taskScopeToneLabel(task.tone)} tone={scopeToneToChip(task.tone)} />
+      </div>
+      <div className="mt-2 flex flex-wrap items-center gap-2 text-[11px] font-medium text-slate-500">
+        <span>{taskStatusLabel(task.status)}</span>
+        <span>{task.evidenceCount} 证据</span>
+        {task.requirementTitle ? <span className="truncate">{task.requirementTitle}</span> : null}
+      </div>
+    </button>
+  );
+}
+
 function ExecutionRunSourcePanel({ run, summary }: { run: ExecutionRunState; summary: PMFlowExecutionSummary }) {
   const prompt = run.prompt;
   return (
@@ -766,12 +962,14 @@ function ExecutionResultCompactPanel({ run }: { run: ExecutionRunState }) {
 function ExecutionRunDecisionPanel({
   run,
   summary,
+  scope,
   capabilities,
   onPrimaryAction,
   onOpenWriteback,
 }: {
   run: ExecutionRunState;
   summary: PMFlowExecutionSummary;
+  scope?: ExecutionScopeBoard | null;
   capabilities?: PMFlowExecutionCapabilitySnapshot | null;
   onPrimaryAction: () => void;
   onOpenWriteback: () => void;
@@ -834,6 +1032,17 @@ function ExecutionRunDecisionPanel({
           </div>
         </section>
 
+        {scope ? (
+          <section className="rounded-lg border border-[#e8eef8] bg-white p-4">
+            <h3 className="text-sm font-semibold text-slate-900">范围缺口</h3>
+            <div className="mt-3 grid grid-cols-2 gap-2">
+              <ExecutionMetric label="待生成包" value={`${scope.totals.waitingPackage}`} tone={scope.totals.waitingPackage ? "warning" : "default"} />
+              <ExecutionMetric label="阻塞任务" value={`${scope.totals.blocked}`} tone={scope.totals.blocked ? "warning" : "default"} />
+            </div>
+            <p className="mt-3 text-xs leading-5 text-slate-500">{scope.nextQueueAction}</p>
+          </section>
+        ) : null}
+
         <section className="rounded-lg border border-[#e8eef8] bg-white p-4">
           <h3 className="text-sm font-semibold text-slate-900">下一步建议</h3>
           <ol className="mt-3 space-y-2">
@@ -1824,6 +2033,240 @@ function StatusChip({ label, tone }: { label: string; tone: "brand" | "success"
   );
 }
 
+function buildExecutionScopeBoard(
+  state: ProjectTruthState,
+  summary: PMFlowExecutionSummary,
+  capabilities?: PMFlowExecutionCapabilitySnapshot | null,
+  requestedTaskId?: string | null
+): ExecutionScopeBoard {
+  const taskItems = buildExecutionScopeTasks(state, summary, capabilities);
+  const requirementMap = new Map(state.requirements.map((item) => [item.id, item]));
+  const groupedTaskIds = new Set<string>();
+  const groups = state.requirements
+    .filter((requirement) => ["confirmed", "in_progress", "done", "released"].includes(requirement.status))
+    .map((requirement) => {
+      const tasks = taskItems.filter((task) => task.requirementId === requirement.id || requirement.linkedTaskIds.includes(task.id));
+      tasks.forEach((task) => groupedTaskIds.add(task.id));
+      return buildRequirementScopeGroup(requirement, tasks);
+    });
+  const unassignedTasks = taskItems.filter((task) => !groupedTaskIds.has(task.id) && (!task.requirementId || !requirementMap.has(task.requirementId)));
+  const allTasks = [...taskItems];
+  const selectedTaskId =
+    (requestedTaskId && allTasks.some((task) => task.id === requestedTaskId) ? requestedTaskId : null) ||
+    (summary.execution.taskId && allTasks.some((task) => task.id === summary.execution.taskId) ? summary.execution.taskId : null) ||
+    firstActionableScopeTask(allTasks)?.id ||
+    allTasks[0]?.id ||
+    null;
+  const totals = {
+    requirements: groups.length,
+    tasks: allTasks.length,
+    running: allTasks.filter((task) => task.tone === "running" || task.tone === "ready").length,
+    waitingPackage: allTasks.filter((task) => task.nextAction.includes("上下文包") || task.nextAction.includes("交接")).length,
+    waitingTest: allTasks.filter((task) => task.nextAction.includes("测试")).length,
+    waitingReview: allTasks.filter((task) => task.nextAction.includes("审查") || task.nextAction.includes("验收")).length,
+    blocked: allTasks.filter((task) => task.tone === "blocked").length,
+  };
+  return {
+    groups,
+    unassignedTasks,
+    allTasks,
+    selectedTaskId,
+    totals,
+    nextQueueAction: buildScopeNextAction(totals),
+  };
+}
+
+function buildExecutionScopeTasks(
+  state: ProjectTruthState,
+  summary: PMFlowExecutionSummary,
+  capabilities?: PMFlowExecutionCapabilitySnapshot | null
+): ExecutionScopeTaskItem[] {
+  const projectTaskMap = new Map(state.tasks.map((task) => [task.id, task]));
+  const devTaskMap = new Map(summary.activeDevTasks.map((task) => [task.id, task]));
+  const taskIds = new Set<string>([
+    ...state.tasks.filter((task) => task.stage === "development" || ["in_progress", "blocked", "done"].includes(task.status)).map((task) => task.id),
+    ...summary.activeDevTasks.map((task) => task.id),
+  ]);
+  if (summary.execution.taskId) taskIds.add(summary.execution.taskId);
+
+  return Array.from(taskIds)
+    .map((taskId) => {
+      const projectTask = projectTaskMap.get(taskId) || null;
+      const devTask = devTaskMap.get(taskId) || null;
+      const task = devTask || (projectTask ? adaptProjectTaskToDevTask(projectTask, state, summary.repoBindings[0] || null) : null);
+      if (!task) return null;
+      return buildScopeTaskItem(task, projectTask, devTask, state, summary, capabilities);
+    })
+    .filter(Boolean) as ExecutionScopeTaskItem[];
+}
+
+function buildScopeTaskItem(
+  task: PMFlowDevTask,
+  projectTask: PMFlowProjectTask | null,
+  devTask: PMFlowDevTask | null,
+  state: ProjectTruthState,
+  summary: PMFlowExecutionSummary,
+  capabilities?: PMFlowExecutionCapabilitySnapshot | null
+): ExecutionScopeTaskItem {
+  const requirement =
+    (projectTask?.requirementId ? state.requirements.find((item) => item.id === projectTask.requirementId) : null) ||
+    state.requirements.find((item) => item.linkedTaskIds.includes(task.id)) ||
+    null;
+  const scopedSummary = createScopedExecutionSummary(summary, state, task.id);
+  const run = buildExecutionRunState(scopedSummary, capabilities);
+  const evidence = scopedSummary.recentProgressEvents.length + (scopedSummary.promptPackage ? 1 : 0);
+  const tone = scopeToneFromRun(run);
+  return {
+    id: task.id,
+    title: task.title,
+    summary: task.description || projectTask?.summary || "任务暂无描述。",
+    status: task.status || projectTask?.status || "unknown",
+    stage: projectTask?.stage || "development",
+    requirementId: requirement?.id || projectTask?.requirementId || task.charter_id || null,
+    requirementTitle: requirement?.title || task.charter_title || null,
+    devTask,
+    projectTask,
+    runStage: run.currentStage.label,
+    nextAction: run.primaryAction.label,
+    evidenceCount: evidence,
+    tone,
+  };
+}
+
+function buildRequirementScopeGroup(
+  requirement: PMFlowRequirement,
+  tasks: ExecutionScopeTaskItem[]
+): ExecutionScopeRequirementGroup {
+  const blocked = tasks.some((task) => task.tone === "blocked");
+  const running = tasks.some((task) => task.tone === "running" || task.tone === "ready");
+  const waiting = tasks.some((task) => task.tone === "waiting");
+  const tone: ExecutionScopeTone = blocked ? "blocked" : running ? "running" : waiting ? "waiting" : tasks.length ? "ready" : "empty";
+  return {
+    id: requirement.id,
+    title: requirement.title,
+    summary: requirement.summary,
+    status: requirement.status,
+    tasks,
+    coverageLabel: tasks.length ? `${tasks.length} 任务` : "未拆任务",
+    tone,
+  };
+}
+
+function firstActionableScopeTask(tasks: ExecutionScopeTaskItem[]) {
+  return (
+    tasks.find((task) => task.tone === "blocked") ||
+    tasks.find((task) => task.tone === "waiting") ||
+    tasks.find((task) => task.tone === "running") ||
+    tasks.find((task) => task.tone === "ready") ||
+    null
+  );
+}
+
+function buildScopeNextAction(totals: ExecutionScopeBoard["totals"]) {
+  if (totals.blocked > 0) return "先处理阻塞";
+  if (totals.waitingPackage > 0) return "生成待执行上下文包";
+  if (totals.waitingTest > 0) return "查看待测试任务";
+  if (totals.waitingReview > 0) return "进入待审查/验收";
+  if (totals.tasks > 0) return "继续跟踪执行证据";
+  return "等待正式开发任务";
+}
+
+function createScopedExecutionSummary(
+  summary: PMFlowExecutionSummary,
+  state: ProjectTruthState,
+  taskId: string
+): PMFlowExecutionSummary {
+  const existingDevTask = summary.activeDevTasks.find((task) => task.id === taskId) || null;
+  const projectTask = state.tasks.find((task) => task.id === taskId) || null;
+  const selectedTask = existingDevTask || (projectTask ? adaptProjectTaskToDevTask(projectTask, state, summary.repoBindings[0] || null) : null);
+  const mergedTasks = selectedTask
+    ? [selectedTask, ...summary.activeDevTasks.filter((task) => task.id !== selectedTask.id)]
+    : summary.activeDevTasks;
+  const scopedEvents = summary.recentProgressEvents.filter((event) => {
+    const eventTaskId = event.task?.id || null;
+    if (!eventTaskId) return taskId === summary.execution.taskId;
+    return eventTaskId === taskId;
+  });
+  const promptMatches = summary.promptPackage && (!summary.promptPackage.task_id || summary.promptPackage.task_id === taskId);
+  const selectedIsCurrent = taskId === summary.execution.taskId;
+  const scopedExecution = selectedIsCurrent
+    ? summary.execution
+    : {
+        ...summary.execution,
+        taskId,
+        status: scopedEvents.length ? summary.execution.status : "queued" as const,
+        prStatus: selectedTask?.code_link?.pr_number ? `PR #${selectedTask.code_link.pr_number}` : "pending",
+        testSummary: "pending",
+        writebackStatus: "pending",
+        commitLog: selectedTask?.code_link?.last_commit_sha ? [selectedTask.code_link.last_commit_sha] : [],
+      };
+  return {
+    ...summary,
+    execution: scopedExecution,
+    promptPackage: promptMatches ? summary.promptPackage : null,
+    activeDevTasks: mergedTasks,
+    recentProgressEvents: scopedEvents,
+  };
+}
+
+function adaptProjectTaskToDevTask(
+  task: PMFlowProjectTask,
+  state: ProjectTruthState,
+  repo: PMFlowRepoBinding | null
+): PMFlowDevTask {
+  const requirement = task.requirementId ? state.requirements.find((item) => item.id === task.requirementId) : null;
+  return {
+    id: task.id,
+    title: task.title,
+    description: task.summary,
+    status: task.status,
+    priority: null,
+    project_id: state.project.id,
+    charter_id: requirement?.id || task.requirementId || null,
+    charter_title: requirement?.title || null,
+    assignee: task.assignee ? { id: task.assignee, name: task.assignee, role: null } : null,
+    prompt_package_count: 0,
+    model_binding: null,
+    code_link: repo
+      ? {
+          id: `${task.id}-repo-link`,
+          status: "pending",
+          branch_name: null,
+          pr_number: null,
+          last_commit_sha: null,
+          repo,
+        }
+      : null,
+  };
+}
+
+function scopeToneFromRun(run: ExecutionRunState): ExecutionScopeTone {
+  if (run.hasFailure || run.currentStage.tone === "blocked") return "blocked";
+  if (run.hasAccepted || run.hasMerged) return "ready";
+  if (run.hasHandoff || run.hasCommit || run.hasPr || run.hasTest) return "running";
+  if (!run.prompt || !run.repo || !run.formalTask) return "waiting";
+  return "ready";
+}
+
+function scopeToneToChip(tone: ExecutionScopeTone): "brand" | "success" | "warning" | "risk" | "muted" {
+  if (tone === "ready") return "success";
+  if (tone === "running") return "brand";
+  if (tone === "waiting") return "warning";
+  if (tone === "blocked") return "risk";
+  return "muted";
+}
+
+function taskScopeToneLabel(tone: ExecutionScopeTone) {
+  const map: Record<ExecutionScopeTone, string> = {
+    ready: "就绪",
+    running: "推进中",
+    waiting: "待补齐",
+    blocked: "阻塞",
+    empty: "空",
+  };
+  return map[tone];
+}
+
 function buildExecutionRunState(
   summary: PMFlowExecutionSummary,
   capabilities?: PMFlowExecutionCapabilitySnapshot | null

frontend/src/features/pmflow-os/store.api.test.tsx

@@ -706,6 +706,40 @@ describe("PMFlowOsProvider real change review actions", () => {
   });
 
   it("renders execution run workspace with staged flow and a single primary action", async () => {
+    const scopedState = realState();
+    scopedState.requirements = [
+      {
+        id: "req-031",
+        title: "REQ-031 执行范围",
+        summary: "本轮需要一起推进的已确认开发需求。",
+        stage: "development",
+        status: "confirmed",
+        linkedTaskIds: ["task-99", "task-extra"],
+      },
+    ];
+    scopedState.tasks = [
+      {
+        id: "task-99",
+        title: "真实任务",
+        summary: "真实任务描述",
+        stage: "development",
+        status: "in_progress",
+        assignee: "Owner",
+        acceptanceCriteria: ["已有执行证据"],
+        requirementId: "req-031",
+      },
+      {
+        id: "task-extra",
+        title: "补充企业权限说明",
+        summary: "同一需求下的另一个开发任务。",
+        stage: "development",
+        status: "confirmed",
+        assignee: "Owner",
+        acceptanceCriteria: ["权限边界明确"],
+        requirementId: "req-031",
+      },
+    ];
+    getTruthStateMock.mockResolvedValueOnce(scopedState);
     getExecutionCapabilitiesMock.mockResolvedValueOnce({
       projectId: "123",
       executionModes: [
@@ -766,6 +800,10 @@ describe("PMFlowOsProvider real change review actions", () => {
 
     expect(await screen.findByText("开发执行流转")).toBeInTheDocument();
     expect(screen.getByText("开发执行 Run 工作台")).toBeInTheDocument();
+    expect(screen.getByText("本轮开发范围")).toBeInTheDocument();
+    expect(screen.getByText("需求执行队列")).toBeInTheDocument();
+    expect(screen.getAllByText("REQ-031 执行范围").length).toBeGreaterThan(0);
+    expect(screen.getByRole("button", { name: /补充企业权限说明/ })).toBeInTheDocument();
     expect(screen.getByText("执行源包")).toBeInTheDocument();
     expect(screen.getByText("AI 副驾")).toBeInTheDocument();
     expect(screen.getAllByText("AI 直接执行").length).toBeGreaterThan(0);
@@ -777,6 +815,9 @@ describe("PMFlowOsProvider real change review actions", () => {
     fireEvent.click(screen.getAllByRole("button", { name: /回写开发 \/ 测试证据/ })[0]);
     expect(screen.getByText("回写执行证据")).toBeInTheDocument();
     expect(screen.getAllByText("运行端已领取，等待 branch、commit、PR 或测试结果回流。").length).toBeGreaterThan(0);
+    fireEvent.click(screen.getByRole("button", { name: /补充企业权限说明/ }));
+    expect(screen.getAllByText("补充企业权限说明").length).toBeGreaterThan(0);
+    expect(screen.getAllByRole("button", { name: /生成上下文包/ }).length).toBeGreaterThan(0);
   });
 
   it("renders Change Review from real projection and does not show demo fallback rows", async () => {