Productize decision assistant archive and context cockpit batch

SESSION.md

@@ -1,8 +1,8 @@
-# ai-exchange Session
+# ai-exchange Session
 
 ## Current Goal
 
-Automation execution-layer PXX remains closed and its known scope stays frozen. Under explicit user direction, the active product phase is a dense `Decision Assistant` operator workspace with grouped recall, broader replay baselines, note templates, quick-save shortcuts, read-only market context, and a continuous operator cockpit. The current run closed the remaining productization gaps from `docs/AI_ICT_实盘交易分析决策助手未完成需求清单_2026-05-14_v1.md`: workflow archive bodies/query/compare, digest archive query/compare, server-persisted focus workspace, and richer digest preference/profile management are now implemented together and aligned with the codebase.
+Automation execution-layer PXX remains closed and its known scope stays frozen. Under explicit user direction, the active product phase is a dense `Decision Assistant` operator workspace with grouped recall, broader replay baselines, note templates, quick-save shortcuts, read-only market context, and a continuous operator cockpit. The previous `鏈畬鎴愰渶姹傛竻鍗昤 has been archived, and the current documentation baseline now uses `docs/AI_ICT_闇€姹傛枃妗ｅ墿浣欏姛鑳戒笌寤舵湡椤规竻鍗昣2026-05-14_v1.md` to distinguish true remaining enhancement work from explicitly deferred real-trading scope.
 
 ## Current Branch
 
@@ -140,6 +140,21 @@ Automation execution-layer PXX remains closed and its known scope stays frozen.
 - latest_decision_assistant_archive_productization_slice: workflow_digest_archive_query_compare
 - latest_decision_assistant_focus_persistence_slice: server_persisted_focus_candidate
 - latest_decision_assistant_profile_management_slice: digest_profiles_and_editable_importance_templates
+- latest_decision_assistant_enhancement_plan_path: docs/AI_ICT_鍒嗘瀽鍔╂墜涓荤嚎澧炲己寮€鍙戣鍒抇2026-05-14_v1.md
+- latest_decision_assistant_preferences_surface_slice: independent_preferences_section
+- latest_decision_assistant_profile_lifecycle_slice: digest_preference_profile_create_rename_delete
+- latest_decision_assistant_preferences_draft_slice: explicit_dirty_save_revert_reset
+- latest_decision_assistant_archive_filter_slice: archive_filter_toolbar_and_query_context
+- latest_decision_assistant_archive_selection_slice: arbitrary_archive_snapshot_compare
+- latest_decision_assistant_archive_pagination_slice: archive_load_more_and_offset_state
+- latest_decision_assistant_focus_evidence_slice: focus_candidate_evidence_chain
+- latest_decision_assistant_focus_digest_slice: focus_related_reminder_recompute
+- latest_decision_assistant_focus_replay_slice: candidate_scoped_replay_rollup
+- latest_decision_assistant_focus_note_slice: structured_focus_note_prefill
+- latest_decision_assistant_macro_provider_slice: macro_news_provider_inventory
+- latest_decision_assistant_chart_registry_slice: multi_provider_chart_reference_registry
+- latest_decision_assistant_exchange_surface_slice: exchange_context_surface_inventory
+- latest_decision_assistant_visual_reference_slice: external_context_visual_reference_cards
 - latest_decision_assistant_summary_status: implemented
 - latest_decision_assistant_change_summary_status: implemented
 - latest_decision_assistant_evidence_status: implemented
@@ -271,7 +286,7 @@ Automation execution-layer PXX remains closed and its known scope stays frozen.
 - latest_p7_selector_choice: continue_periodic_analysis_decision_monitoring_with_maturity_rollup
 - latest_p14_slice: webhook_transport_retry
 - latest_p15_pxx_tasklist_status: ready
-- latest_p15_pxx_tasklist_path: docs/AI_ICT_自动化执行层_P15-PXX闭环任务清单_v1.md
+- latest_p15_pxx_tasklist_path: docs/AI_ICT_鑷姩鍖栨墽琛屽眰_P15-PXX闂幆浠诲姟娓呭崟_v1.md
 - latest_p15_slice: alert_noise_policy_matrix
 - latest_alert_noise_policy_status: implemented
 - latest_alert_noise_policy_route_status: visible_existing_quality_summary
@@ -525,7 +540,7 @@ Automation execution-layer PXX remains closed and its known scope stays frozen.
 - Added bounded webhook transport retry inside `AlertDispatcherService._send` so transient outbound failures can be retried before the alert is marked failed.
 - Added transport-retry observability in dispatcher payloads, delivery summary fields, and runtime `alert_dispatch` details.
 - Added focused unit tests for the service, script, API routing, and runtime task.
-- Added `docs/AI_ICT_自动化执行层_P15-PXX闭环任务清单_v1.md` to define a finite continuation queue from P15 through PXX.
+- Added `docs/AI_ICT_鑷姩鍖栨墽琛屽眰_P15-PXX闂幆浠诲姟娓呭崟_v1.md` to define a finite continuation queue from P15 through PXX.
 - Updated the state anchors so future continuation starts at P15 instead of inventing new unbounded phases.
 - Added alert noise policy matrix settings for external minimum severity, workbench-only alert types, suppressed duty modes, and suppressed operator modes.
 - Extended `DeliveryPolicyService` with `evaluate_alert_delivery_policy()` so alert status, silent window, alert type, severity, channel, duty mode, operator mode, and cooldown return a structured policy decision.
@@ -688,11 +703,21 @@ Automation execution-layer PXX remains closed and its known scope stays frozen.
 - Latest E8 validation passed: focused exchange/provider tests, 71 related unit/API/static tests, 717 full unit tests, 73 script static checks, deploy completed, remote ops smoke passed 9 checks, and remote field smoke confirmed `exchange_status/announcements = idle`, `cache_state = empty_fallback`, and `chart_stale_source_count = 0`.
 - Latest E9/B1 validation passed: 72 related unit/API/static tests, 720 full unit tests, 73 script static checks, `node --check src/web/app.js`, remote deploy completed, remote ops smoke passed 9 checks, and remote field smoke confirmed `chart_url`, `symbol=BTC-USDT`, `interval=1m`, adapter mode `live_reference/empty_fallback`, and `unattended_harness = passed`.
 - Latest W6/F27/F28/H29 validation passed: 70 focused route/operator/static tests, 722 full unit tests, 73 script static checks, `node --check src/web/app.js`, deploy smoke passed 11 endpoints, remote ops smoke passed 9 checks, and remote field smoke confirmed digest/workflow archive query+compare routes, profile route visibility, and non-blocking assistant focus route.
+- Latest A1/A2/A3/A4 validation passed: 71 focused API/operator/static tests, 723 full unit tests, 73 script static checks, `node --check src/web/app.js`, deploy smoke passed 11 endpoints, remote ops smoke passed 9 checks, and remote Phase A smoke confirmed profile create/save/review-packet profile switch/rename/delete closure.
+- Latest B1/B2/B3/B4 validation passed: 71 focused API/operator/static tests, 723 full unit tests, 73 script static checks, `node --check src/web/app.js`, deploy smoke passed 11 endpoints, remote ops smoke passed 9 checks, and remote Phase B smoke confirmed archive filters, arbitrary digest/workflow snapshot compare, and load-more compatible query state.
+- Latest C1/C2/C3/C4 validation passed: 71 focused API/operator/static tests, 723 full unit tests, 73 script static checks, `node --check src/web/app.js`, deploy smoke passed 11 endpoints, remote ops smoke passed 9 checks, and remote Phase C smoke confirmed persisted focus drives evidence, reminders, replay scope, and structured note prefill.
+- Latest D1/D2/D3/D4 validation passed: 23 focused provider/operator/static tests, 723 full unit tests, 73 script static checks, `node --check src/web/app.js`, deploy smoke passed 11 endpoints, remote ops smoke passed 9 checks, and remote Phase D smoke confirmed macro provider inventory, chart provider registry, exchange surface inventory, and visual reference cards on the live host.
+- Latest E1 validation passed: 74 focused workflow/operator/static tests, 726 full unit tests, 73 script static checks, `node --check src/web/app.js`, deploy smoke passed 11 endpoints, remote ops smoke passed 9 checks, and remote Phase E smoke confirmed workflow archive operations summary, day/session groups, multi-day drift review, and deployed `workflow operations` / `workflow drift` / `Day Group` / `Session Group` UI strings on the live host.
+- Latest B5 validation passed: 72 focused API/static/operator tests, 726 full unit tests, 73 script static checks, `node --check src/web/app.js`, live deploy completed, remote ops smoke passed 9 checks, and remote B5 smoke confirmed digest/workflow archive `query_text` search, digest/workflow export bundles, and deployed `Search Text` / `Export Archive Results` / `Export Workflow Results` UI strings on the live host.
+- Latest W7 validation passed: 74 focused workflow/API/static/operator tests, 726 full unit tests, 73 script static checks, `node --check src/web/app.js`, live deploy completed, remote ops smoke passed 9 checks, and remote W7 smoke confirmed workflow archive bridge summary, handoff export bundle summary, and deployed `Replay Bridge` / `Note Bridge` / `Handoff Bridge` / `Prefill Replay Note` UI strings on the live host.
+- Latest E10 validation passed: 16 focused external-context/static tests, 726 full unit tests, 73 script static checks, `node --check src/web/app.js`, live deploy completed, remote ops smoke passed 9 checks, and remote E10 smoke confirmed provider health summary/cards, visual evidence bridge summary, and deployed `Open Visual Bridge` / `Open Target` / `Open Embed` UI strings on the live host.
+- Latest O1 validation passed: 19 focused archive/static/operator tests, 729 full unit tests, 73 script static checks, `node --check src/web/app.js`, live deploy completed, remote ops smoke passed 9 checks, and remote O1 smoke confirmed digest/workflow long-range trend summaries, baseline pin starter fields, and deployed `archive trend` / `archive baseline` / `Pin Baseline` / `Compare To Baseline` UI strings on the live host.
+- Refreshed `docs/AI_ICT_??????????????????????2026-05-14_v1.md` after closing Phases A-E; the active remaining mainline backlog is now archive deep retrieval follow-up, external context deep visual evidence, workflow archive deep bridge extensions, and archive long-range trend deepening.
 - Real-live order submission remains intentionally disabled because the OKX order adapter is still not a real-live submitter.
 
 ## Exact Next Step
 
-The `未完成需求清单` alignment batch is now the active closure target: deploy the current Decision Assistant archive/focus/profile productization work, smoke the new archive/focus routes, and then treat any further work as non-blocking enhancement depth rather than unfinished baseline scope. PXX remains closed and you still must not create P25/P26 automatically or enable real trading.
+Use `docs/AI_ICT_???????????????????2026-05-14_v1.md` as the completed enhancement record for Phases A-E. The active backlog source is now `docs/AI_ICT_??????????????????????2026-05-14_v1.md`. `B5 archive keyword search / export starter slice`, `W7 workflow archive replay / handoff / export bridge starter slice`, `E10 external context visual evidence / provider health starter slice`, and `O1 archive long-range trend / baseline pin starter slice` are now closed. Start from `B6 archive deep retrieval / structured export follow-up starter slice`, keep work on the analysis-assistant mainline only, and continue treating real-trading scope as deferred. PXX remains closed and you still must not create P25/P26 automatically or enable real trading.
 
 ## Validation Commands Used Recently
 
@@ -711,6 +736,12 @@ D:\ai-exchange\.venv\Scripts\python.exe -m unittest discover -s tests/unit
 node --check src/web/app.js
 D:\ai-exchange\.venv\Scripts\python.exe scripts\deploy_with_password.py --host 192.168.3.90 --user root --port 22 --password <provided>
 D:\ai-exchange\.venv\Scripts\python.exe scripts\run_remote_ops_smoke.py --base-url http://192.168.3.90:8000 --attempts 2 --timeout-seconds 20
+D:\ai-exchange\.venv\Scripts\python.exe -m unittest tests.unit.test_workflow_archive_projection tests.unit.test_operator_workspace_service tests.unit.test_api_static_server tests.unit.test_api_server
+D:\ai-exchange\.venv\Scripts\python.exe scripts\run_script_static_checks.py
+D:\ai-exchange\.venv\Scripts\python.exe -m unittest discover -s tests/unit
+node --check src/web/app.js
+D:\ai-exchange\.venv\Scripts\python.exe scripts\deploy_with_password.py --host 192.168.3.90 --user root --port 22 --password <provided>
+D:\ai-exchange\.venv\Scripts\python.exe scripts\run_remote_ops_smoke.py --base-url http://192.168.3.90:8000 --attempts 2 --timeout-seconds 20
 D:\ai-exchange\.venv\Scripts\python.exe -m unittest tests.unit.test_visual_review_service tests.unit.test_run_visual_review tests.unit.test_trading_runtime_service tests.unit.test_api_and_main tests.unit.test_api_server
 D:\ai-exchange\.venv\Scripts\python.exe scripts\run_script_static_checks.py
 D:\ai-exchange\.venv\Scripts\python.exe -m unittest tests.unit.test_run_until_blocked tests.unit.test_monitoring_alert_rollout tests.unit.test_runtime_script_json

TASKS.md

@@ -103,6 +103,29 @@
 
 ## Next Up
 
+- [x] B5 archive keyword search / export starter slice
+- [x] W7 workflow archive replay / handoff / export bridge starter slice
+- [x] E10 external context visual evidence / provider health starter slice
+- [x] O1 archive long-range trend / baseline pin starter slice
+- [ ] B6 archive deep retrieval / structured export follow-up starter slice
+
+- [x] A1 independent preferences surface / dedicated settings section
+- [x] A2 preference profile create / duplicate / rename / delete
+- [x] A3 global and template threshold editor / reset defaults
+- [x] A4 dirty draft / explicit save / revert
+- [x] B1 archive deep retrieval / arbitrary snapshot compare
+- [x] B2 archive filter toolbar / query context / date-range retrieval
+- [x] B3 archive load-more / offset-state browser
+- [x] B4 digest and workflow compare drill-down
+- [x] C1 focus-aware evidence chain
+- [x] C2 focus-aware digest reminder recompute
+- [x] C3 candidate-scoped replay learning rollup
+- [x] C4 structured focus note prefill / submit context
+- [x] D1 multi-provider chart reference registry
+- [x] D2 macro/news provider inventory enrichment
+- [x] D3 exchange context surface inventory
+- [x] D4 external context visual reference cards
+- [x] E1 workflow archive operations view / multi-day drift review
 - [x] W6 workflow archive bodies / query / compare productization slice complete
 - [x] F27 digest archive query / compare productization slice complete
 - [x] F28 digest profile management / editable importance template slice complete

docs/AI_ICT_全自动开发执行流程_v1.md

@@ -1,4 +1,4 @@
-# AI ICT 全自动开发执行流程 v1
+# AI ICT 全自动开发执行流程 v1
 
 ## 1. 目标
 
@@ -98,15 +98,15 @@
 
 - `docs/archive/AI_ICT_下一阶段开发计划_v1.md`
 - `docs/archive/AI_ICT_下一阶段开发任务清单_v1.md`
-- `docs/AI_ICT_TradingAgents_P0自动任务清单_v1.md`
+- `docs/archive/AI_ICT_TradingAgents_P0自动任务清单_v1.md`
 - `docs/AI_ICT_TradingAgents_P1自动任务清单_v1.md`
 - `docs/AI_ICT_TradingAgents_P2自动任务清单_v1.md`
 - `docs/AI_ICT_TradingAgents_P3自动任务清单_v1.md`
 - `docs/AI_ICT_TradingAgents_P4自动任务清单_v1.md`
 - `docs/AI_ICT_TradingAgents_P5自动任务清单_v1.md`
 - `docs/AI_ICT_TradingAgents_P6自动任务清单_v1.md`
-- `docs/AI_ICT_TradingAgents_P0-P6合理性审查与优化结论_v1.md`
-- `docs/AI_ICT_research_council脚手架与数据表草案_v1.md`
+- `docs/archive/AI_ICT_TradingAgents_P0-P6合理性审查与优化结论_v1.md`
+- `docs/archive/AI_ICT_research_council脚手架与数据表草案_v1.md`
 
 ## 4.3 真源原则
 
@@ -364,7 +364,7 @@ npm run visual:snapshots
 
 必要时同步更新：
 
-- `docs/AI_ICT_TradingAgents融合实施方案_最终版.md`
+- `docs/archive/AI_ICT_TradingAgents融合实施方案_最终版.md`
 - 当前阶段相关任务清单 / 技术设计文档
 
 必须写回的内容：
@@ -688,7 +688,7 @@ no_op_healthy 时停止
 
 正式进入自动化执行层开发的条件：
 
-1. `docs/AI_ICT_TradingAgents_P0-P6合理性审查与优化结论_v1.md` 已存在
+1. `docs/archive/AI_ICT_TradingAgents_P0-P6合理性审查与优化结论_v1.md` 已存在
 2. `SESSION.md` / `TASKS.md` 已写明 `automation_flow_p0_p6_review_closed`
 3. P6 governance tick 最近一次结果为 `no_op_healthy`
 4. 文档一致性测试通过
@@ -740,7 +740,7 @@ P7 必须继续遵守：
 P7 任务清单以以下文档为准：
 
 ```text
-docs/AI_ICT_自动化执行层_P7监控成熟度与晋级路由_v1.md
+docs/archive/AI_ICT_自动化执行层_P7监控成熟度与晋级路由_v1.md
 ```
 
 P14 之后的自动化执行层连续开发，以以下有限清单为准：
@@ -827,3 +827,4 @@ docs/AI_ICT_自动化执行层_P15-PXX闭环任务清单_v1.md
 换句话说：
 
 > **以后该项目默认全程采用自动化流程推进，除非遇到真实外部阻塞或必须用户拍板的产品决策。**
+

docs/AI_ICT_分析助手主线增强开发计划_2026-05-14_v1.md

@@ -0,0 +1,472 @@
+# AI_ICT 分析助手主线增强开发计划 2026-05-14 v1
+
+## 1. 目标
+
+本计划只围绕当前“分析助手”主线推进，不展开真实交易、OKX live submit、默认自动下单、live adapter 开闸。
+
+本轮增强的目标是把当前已经可用的 `Decision Assistant`，从“有基础闭环的工作台”进一步推进为：
+
+- 可配置
+- 可检索
+- 可对比
+- 可聚焦
+- 可连续日常使用
+
+重点只覆盖以下 5 个方向：
+
+1. 独立偏好设置面
+2. archive 深度检索与任意对比
+3. candidate-scoped 连续工作区重算
+4. 外部上下文 provider 与视觉化增强
+5. workflow archive 运营化能力
+
+## 2. 边界
+
+本计划明确不做：
+
+- 真实交易执行
+- OKX live submit
+- 默认自动下单
+- risk governor / trust bucket / operator confirmation protocol 逻辑变更
+- execution_allowed / real_live_ready 开闸
+
+本计划只做：
+
+- 分析助手体验增强
+- 只读数据组织增强
+- 历史检索与比较增强
+- 工作台聚焦与连贯性增强
+
+## 3. 当前基线
+
+当前代码已具备的基础：
+
+- digest preferences 持久化
+- digest history archive / workflow history archive
+- latest-two compare 基础能力
+- focus candidate 持久化基础版
+- continuous workspace / loop queue / watchlist / replay / notes 入口
+- 外部上下文基础卡片
+- external chart read-only provider 基础版
+
+因此本轮开发不是从零搭建，而是做“产品层增强”。
+
+## 4. 总体推进顺序
+
+建议按以下顺序推进：
+
+### Phase A
+
+- 独立偏好设置面
+
+### Phase B
+
+- archive 深度检索与任意对比
+
+### Phase C
+
+- candidate-scoped 连续工作区重算
+
+### Phase D
+
+- 外部上下文 provider 与视觉化增强
+
+### Phase E
+
+- workflow archive 运营化能力
+
+原因：
+
+- A 和 B 能最快提升日常可用性
+- C 能把现有 cockpit 从“看板”推进成“工作区”
+- D 和 E 更偏增强深度，适合在前 3 项稳定后推进
+
+## 5. Phase A: 独立偏好设置面
+
+### 5.1 目标状态
+
+从当前“通过内联按钮和 action bar 保存偏好”，升级到独立的 `Preferences` 工作区。
+
+用户应能：
+
+- 查看当前 profile
+- 新建 / 复制 / 重命名 / 删除 profile
+- 调整 digest mode
+- 调整 preferred history kind
+- 调整 global importance threshold
+- 单独编辑 importance template threshold
+- 配置 history window
+- 明确保存 / 重置 / 撤销
+
+### 5.2 交付范围
+
+后端：
+
+- profile 管理 API
+- profile 新建 / 复制 / 删除
+- payload schema 稳定化
+- profile label 唯一性与回退规则
+
+前端：
+
+- 新增独立 `Preferences` section
+- profile selector
+- profile actions
+- threshold editor
+- save / reset / duplicate / delete controls
+- dirty state 提示
+
+数据层：
+
+- 继续复用 `decision_assistant_preferences`
+- 增补 payload 规范：
+  - `profile_label`
+  - `history_window_hours`
+  - `importance_templates`
+  - `focus_candidate_id`
+  - 后续可拓展字段占位
+
+### 5.3 建议切片
+
+#### A1 profile list + selector
+
+- `GET /trading/alerts/digest-preferences/profiles`
+- 前端独立展示当前 profile
+- UI 不再只靠 action bar
+
+#### A2 profile create / duplicate / rename / delete
+
+- 后端加管理 route
+- 前端支持 profile 级管理动作
+
+#### A3 threshold editor
+
+- global threshold
+- template threshold
+- reset to default
+
+#### A4 dirty state + explicit save/revert
+
+- 未保存提示
+- 手动保存
+- 取消修改
+
+### 5.4 验收标准
+
+- 不进入 digest 主卡片，也能完整管理偏好
+- 新 profile 能创建并立即生效
+- 修改 template threshold 后刷新仍保留
+- 删除 profile 不影响默认 `local_operator`
+
+### 5.5 focused validation
+
+- repository / API unit tests
+- `test_api_server`
+- `test_operator_workspace_service`
+- `test_api_static_server`
+- `node --check src/web/app.js`
+
+## 6. Phase B: archive 深度检索与任意对比
+
+### 6.1 目标状态
+
+从当前“latest-two compare + 基础 limit/offset”，升级到真正可用的 archive 工作流。
+
+用户应能：
+
+- 按 profile 检索
+- 按日期范围检索
+- 按 digest mode / history kind / workflow phase 检索
+- 从列表中任意选择两条 snapshot 做 compare
+- 查看 compare 差异摘要
+
+### 6.2 交付范围
+
+后端：
+
+- digest archive query 扩展
+- workflow archive query 扩展
+- compare route 支持任意 snapshot id 对
+- 支持时间过滤与 profile 过滤
+
+前端：
+
+- archive filter toolbar
+- archive result list
+- dual-selection compare UI
+- clear filter / clear compare
+- compare summary card
+
+### 6.3 建议切片
+
+#### B1 query 扩展
+
+- created_after / created_before
+- profile_key
+- workflow phase / artifact kind
+
+#### B2 archive result browser
+
+- 筛选器
+- 分页或“加载更多”
+- 当前检索上下文可见
+
+#### B3 arbitrary compare
+
+- 勾选两条 snapshot
+- compare route 取差异
+- compare summary / changed fields / count delta
+
+#### B4 compare drill-down
+
+- digest：title/body/reminder delta
+- workflow：phase brief/body delta
+
+### 6.4 验收标准
+
+- 可以不依赖“最新两条”，手动选任意两条 archive compare
+- 日期过滤可稳定返回结果
+- digest 和 workflow 两条线都能检索和对比
+
+### 6.5 focused validation
+
+- repository query tests
+- archive compare tests
+- API route tests
+- dashboard JS syntax
+
+## 7. Phase C: candidate-scoped 连续工作区重算
+
+### 7.1 目标状态
+
+从当前“focus persistence + 局部联动”，升级到真正围绕一个 candidate 重算工作区。
+
+当用户选中一个 focus candidate 后：
+
+- Evidence 只展示该 candidate 相关证据链
+- Digest 只展示与该 candidate 强相关提醒
+- Replay 优先展示该 candidate 对应 setup/session 学习面
+- Notes 默认带入该 candidate 的结构化上下文
+
+### 7.2 交付范围
+
+后端：
+
+- candidate-scoped evidence builder
+- candidate-scoped digest reminder builder
+- candidate-scoped replay learning rollup
+- focus-aware review entry defaults
+
+前端：
+
+- focus candidate 明确可见
+- clear focus / switch focus
+- 焦点切换后 section 结果一致刷新
+
+### 7.3 建议切片
+
+#### C1 focus-aware evidence chain
+
+- evidence 过滤与重算
+- evidence source path 继续保持可解释
+
+#### C2 focus-aware digest
+
+- reminders 以当前 focus candidate 为第一优先级
+- 当前 focus 不相关提醒降级
+
+#### C3 focus-aware replay
+
+- setup / session / market state 统计优先围绕当前 focus
+
+#### C4 focus-aware notes
+
+- 新 note 默认带入：
+  - candidate_id
+  - session_code
+  - model_code
+  - focus title
+
+### 7.4 验收标准
+
+- 切换 focus candidate 后，Evidence / Digest / Replay / Notes 明显随之变化
+- focus 状态刷新页面后仍保留
+- “清除 focus” 后退回默认全局视角
+
+### 7.5 focused validation
+
+- `test_operator_workspace_service`
+- `test_api_server`
+- front-end interaction smoke
+- remote field smoke for focus-aware sections
+
+## 8. Phase D: 外部上下文 provider 与视觉化增强
+
+### 8.1 目标状态
+
+从当前“有基础 provider 和 reference 卡片”，升级到更丰富的外部上下文层。
+
+用户应能：
+
+- 查看多个外部上下文源
+- 知道当前 source freshness / status / degrade state
+- 在需要时看到可视化的外部 context 证据
+
+### 8.2 交付范围
+
+后端：
+
+- 多 provider external chart source 抽象
+- 宏观/新闻 provider 扩展
+- 交易所状态页 / 异常页 source 扩展
+- 外部 context visual evidence 元数据
+
+前端：
+
+- provider list
+- source state / freshness / fallback tag
+- chart reference card 增强
+- 可选 screenshot / embed 入口
+
+### 8.3 建议切片
+
+#### D1 multi-provider registry
+
+- `binance_public_chart_reference` 之外增加 provider 抽象
+- provider selection / fallback order
+
+#### D2 macro/news provider enrichment
+
+- 丰富事件类型
+- provider freshness 统一摘要
+
+#### D3 exchange anomaly surface enrichment
+
+- 多交易所只读状态页
+- source cards 中统一展示
+
+#### D4 visual evidence extension
+
+- 外部 chart/image 仅作为只读可视证据
+- 不改变主分析链路
+
+### 8.4 验收标准
+
+- external context 至少支持多 provider registry
+- UI 能显示 provider 状态差异
+- screenshot / embed 失败不会阻塞主分析
+
+### 8.5 focused validation
+
+- provider tests
+- exchange status tests
+- workbench static smoke
+- remote route field smoke
+
+## 9. Phase E: workflow archive 运营化能力
+
+### 9.1 目标状态
+
+从当前“能看 archive 条目”，升级到“可用于盘前/盘中/盘后运营回看”的 archive 视图。
+
+用户应能：
+
+- 按日历日回看 workflow
+- 按 session 回看 workflow
+- 查看某天完整 pre/intraday/post 流转
+- 对比多天 workflow drift
+
+### 9.2 交付范围
+
+后端：
+
+- day/session grouped workflow archive query
+- workflow daily summary builder
+- workflow drift compare summary
+
+前端：
+
+- 按天浏览
+- 按 session 浏览
+- workflow timeline summary
+- multi-day compare card
+
+### 9.3 建议切片
+
+#### E1 grouped archive query
+
+- day bucket
+- session bucket
+- item counts / status counts
+
+#### E2 daily summary view
+
+- 某天 pre/intraday/post 摘要串联
+
+#### E3 drift compare
+
+- 多日 artifact count / pending/update drift
+- 多日 body 变化摘要
+
+#### E4 export / replay handoff
+
+- 为后续复盘导出预留结构
+
+### 9.4 验收标准
+
+- 用户能按天看 workflow
+- 用户能看出多天之间 workflow 差异
+- archive 不再只是条目列表，而是运营回看面
+
+### 9.5 focused validation
+
+- workflow history service tests
+- API tests
+- front-end render tests
+- remote smoke for grouped workflow archive
+
+## 10. 推荐排期
+
+如果按最稳的开发节奏推进，建议分 5 轮：
+
+### Round 1
+
+- Phase A1-A4
+
+### Round 2
+
+- Phase B1-B4
+
+### Round 3
+
+- Phase C1-C4
+
+### Round 4
+
+- Phase D1-D4
+
+### Round 5
+
+- Phase E1-E4
+
+## 11. 每轮统一闭环要求
+
+每一轮都必须完成：
+
+1. 锚点读取
+2. 最小闭环切片实现
+3. focused validation
+4. 必要质量门
+5. 必要时远端 deploy/smoke
+6. 回写 `SESSION.md / TASKS.md`
+
+## 12. 最终结论
+
+这 5 个方向不是新的大重构，而是围绕当前分析助手主线的连续增强计划。
+
+推荐默认从 `独立偏好设置面` 开始，原因是：
+
+- 对日常使用提升最大
+- 风险最低
+- 与后续 archive / focus 深化天然衔接
+- 不触碰延期项与实盘权限边界

docs/AI_ICT_需求文档剩余功能与延期项清单_2026-05-14_v1.md

@@ -0,0 +1,173 @@
+# AI_ICT 需求文档剩余功能与延期项清单 2026-05-14 v1
+
+## 1. 对照范围
+
+本清单基于以下仍然有效的需求文档与当前代码状态整理：
+
+1. `docs/AI_ICT_实盘交易分析决策助手下一阶段开发方案_v1.md`
+2. `docs/AI_ICT_实盘辅助决策系统开发方案_v2.md`
+3. `docs/AI_ICT_实盘交易分析系统开发计划_v1.md`
+4. `docs/ICT交易全书_出版终校版.md`
+5. `SESSION.md`
+6. `TASKS.md`
+
+说明：
+- 本文只列“截至当前代码仍未完全实现”的内容。
+- 已完成并已闭环的能力不再重复列为待办。
+- 与真实交易、OKX live submit、默认自动下单、real live readiness 开闸相关内容，继续单列为延期项，不纳入当前分析助手主线 backlog。
+
+## 2. 已完成且已闭环的主线能力
+
+以下能力已经完成到可用闭环，不再视为“当前缺口”：
+
+- Decision Assistant dense cockpit 主界面
+- unified watchlist / evidence / digest / replay / notes 连续工作区基础能力
+- digest 偏好持久化、profile 管理、独立 preferences section
+- digest archive / workflow archive 查询、load-more、任意双选 compare
+- pre-market / intraday / post-trade workflow rhythm 与 artifact persistence
+- candidate-scoped focus workspace 基础重算
+- replay learning / weekly trend / attribution rollup
+- external context provider inventory、exchange surface inventory、visual reference cards
+- workflow archive operations view、day/session grouping、multi-day drift review
+
+结论：
+- 之前 Phase A-E 对应的 5 个增强方向已经全部闭环完成。
+- 这份文档现在需要承担的是“下一批 backlog 刷新”，而不是继续重复 A-E。
+
+## 3. 当前仍未完全实现的功能
+
+### 3.1 archive 全文检索与导出能力
+
+当前状态：
+
+- digest archive / workflow archive 已支持 profile、kind、mode、phase、date range、limit、offset。
+- 已支持任意两条 snapshot compare。
+- 已支持 UI 浏览、load-more、compare drill-down。
+- 已支持 `query_text` 关键词检索。
+- 已支持当前过滤结果与当前 compare 结果的 JSON export starter bundle。
+
+仍缺：
+
+- 跨 profile archive 检索
+- 更细颗粒度的结构化导出格式
+- 常用 archive 查询预设保存
+
+结论：
+
+- `archive 查询、compare、关键词检索与基础导出已完成`
+- `archive 深化检索与运营化导出能力未完成`
+
+### 3.2 external context 视觉证据增强层
+
+当前状态：
+
+- 已有 macro/news provider inventory、chart provider registry、exchange context source inventory。
+- 已有 visual reference cards 和只读外部 chart reference。
+- 已有 provider health cards、provider health summary、visual evidence bridge summary。
+- cockpit 已支持 `Open Visual Bridge` / `Open Target` / `Open Embed` 只读打开动作。
+
+仍缺：
+
+- 外部 chart embed / screenshot capture / visual evidence 化
+- 更丰富的宏观 / 新闻 / 交易所异常 provider 扩展
+- external context 与 evidence chain 的更深联动
+
+结论：
+
+- `external context inventory、provider health、visual bridge starter 已完成`
+- `external context 的深层视觉证据增强层未完成`
+
+### 3.3 workflow archive 回放与导出桥接
+
+当前状态：
+
+- workflow archive 已能持久化正文、浏览、compare、按 day/session group 汇总。
+- 已有 operations summary 和 multi-day drift review。
+- 已有 archive snapshot 到 `Setup Replay / Review Entry / Workflow` 的 bridge action。
+- workflow export bundle 已带基础 `bridge_summary / handoff_bundle_summary`。
+
+仍缺：
+
+- day/session 级 artifact bundle 导出
+- workflow archive 到 operator handoff packet 的快速生成
+- archive 对比结果回写为结构化复盘记录
+
+结论：
+
+- `workflow archive 运营化浏览层与 starter bridge 已完成`
+- `workflow archive 到 replay / handoff / export 的深层桥接未完成`
+
+### 3.4 archive 运营化长期视图
+
+当前状态：
+
+- 当前 archive 视图已经覆盖近期 snapshot 浏览、day/session grouping、latest drift compare。
+- 已有 7d / 30d long-range trend starter summary。
+- 已有 baseline pin starter fields 与 compare prompt。
+
+仍缺：
+
+- baseline pin / compare baseline 的持久化保存
+- digest / workflow 的长期趋势联动视图深化
+- 面向运营回看的筛选、趋势、导出统一入口
+
+结论：
+
+- `短周期 archive 运维视图与长期趋势 starter 已完成`
+- `长期趋势深化与运营统一入口未完成`
+
+## 4. 明确延期项
+
+以下内容仍然属于延期项，不纳入当前分析助手主线：
+
+### 4.1 真实交易执行
+
+- OKX live submit
+- 默认真实自动下单
+- live order routing / bracket / OCO 正式投产
+- execution_allowed / real_live_ready 开闸
+
+### 4.2 真实账户与实盘同步深化
+
+- 更深的 broker read-only sync 扩展
+- 围绕真实 fill 的 reconciliation 运营闭环
+- 真实仓位驱动的半自动执行协同
+
+### 4.3 live adapter 最终放量阶段
+
+- real-live readiness 最终验收
+- operator confirmation protocol 的真实下单阶段验收
+
+说明：
+- 这些不是“忘了做”，而是当前阶段明确延后。
+
+## 5. 建议的下一批 backlog 顺序
+
+如果继续围绕“分析助手”主线推进，建议顺序如下：
+
+1. `archive 深化检索能力`
+2. `external context 深层视觉证据增强`
+3. `workflow archive 深层桥接`
+4. `archive 长期趋势深化`
+
+## 6. 建议的下一刀最小闭环切片
+
+推荐下一轮直接从这一刀开始：
+
+`B6 archive deep retrieval / structured export follow-up starter slice`
+
+建议范围：
+
+- 为 digest / workflow archive 增加跨 profile 检索 follow-up
+- 补结构化 export format 与 compare/export schema 深化
+- 增加 archive 查询 preset starter fields
+- 保持只读，不改变交易权限与主分析决策边界
+- 完成 focused validation + 必要 deploy/smoke + 回写 `SESSION.md / TASKS.md`
+
+## 7. 最终结论
+
+截至当前代码状态：
+
+- A-E 计划已经全部完成，不该再作为主 backlog 重复推进。
+- 当前真正未完成的，已经从“基础产品化”切换到“archive 深化、external evidence 强化、运营长期视图”这一级。
+- 真实交易、live adapter、下单权限相关内容继续保持延期，不应混回当前分析助手主线。

scripts/run_monitoring_alert_rollout.py

@@ -256,9 +256,9 @@ PHASES = [
             "N4-2 仅做代码预留点",
         ],
         "files": [
-            "docs/AI_ICT_盯盘告警架构图_v1.md",
-            "docs/AI_ICT_盯盘告警技术设计稿_v1.md",
-            "docs/AI_ICT_盯盘告警开发任务清单_v1.md",
+            "docs/archive/AI_ICT_盯盘告警架构图_v1.md",
+            "docs/archive/AI_ICT_盯盘告警技术设计稿_v1.md",
+            "docs/archive/AI_ICT_盯盘告警开发任务清单_v1.md",
             "src/services/decision/decision_packet_service.py",
             "src/services/runtime/trading_runtime_service.py",
         ],
@@ -1329,3 +1329,4 @@ def main(env=None) -> int:
 
 if __name__ == "__main__":
     raise SystemExit(main())
+

scripts/run_until_blocked.py

@@ -3366,7 +3366,7 @@ def render_session(completed_stages: list[str], current_stage: Optional[str], la
             "  - `TASKS.md`",
             "- 当前执行协议说明保留在：",
             "  - `docs/AI_ICT_全自动开发执行流程_v1.md`",
-            "- `docs/AI_ICT_TradingAgents_P0-P6合理性审查与优化结论_v1.md` 仅作为已完成冻结决策的历史治理参考，不再是每次启动必读锚点。",
+            "- `docs/archive/AI_ICT_TradingAgents_P0-P6合理性审查与优化结论_v1.md` 仅作为已完成冻结决策的历史治理参考，不再是每次启动必读锚点。",
             "- 默认不再等待用户中途确认",
             "- 每阶段必须完成：",
             "  - 实现",
@@ -3469,3 +3469,4 @@ def main() -> int:
 
 if __name__ == "__main__":
     raise SystemExit(main())
+

src/api/alerts.py

@@ -1,5 +1,8 @@
 from __future__ import annotations
 
+from datetime import datetime, timezone
+import re
+
 from src.repositories.alert_delivery_repository import AlertDeliveryRepository
 from src.repositories.decision_assistant_repository import DecisionAssistantRepository
 from src.repositories.trade_plan_candidate_repository import TradePlanCandidateRepository
@@ -9,6 +12,13 @@ from src.services.alerts import (
     AlertSilentWindowService,
     OperatorDigestService,
 )
+from src.services.operator.archive_long_range_projection import (
+    build_digest_archive_baseline_summary,
+    build_digest_archive_long_range_summary,
+    build_workflow_archive_baseline_summary,
+    build_workflow_archive_long_range_summary,
+)
+from src.services.operator.workflow_archive_projection import build_workflow_archive_operations_view
 
 
 def dispatch_pending_alerts(limit: int = 50) -> dict:
@@ -79,11 +89,11 @@ def list_operator_digest_preference_profiles(instrument_id: int) -> dict:
                 "updated_ts": row.get("updated_ts"),
             }
         )
-    if not items:
-        default = _default_preference_payload(
-            instrument_id=instrument_id,
-            profile_key=DecisionAssistantRepository.DEFAULT_PROFILE_KEY,
-        )
+    default = _default_preference_payload(
+        instrument_id=instrument_id,
+        profile_key=DecisionAssistantRepository.DEFAULT_PROFILE_KEY,
+    )
+    if not any(item.get("profile_key") == DecisionAssistantRepository.DEFAULT_PROFILE_KEY for item in items):
         items.append(
             {
                 "profile_key": default.get("profile_key"),
@@ -103,6 +113,141 @@ def list_operator_digest_preference_profiles(instrument_id: int) -> dict:
     }
 
 
+def create_operator_digest_preference_profile(
+    instrument_id: int,
+    *,
+    profile_key: str | None = None,
+    profile_label: str | None = None,
+    source_profile_key: str | None = None,
+    digest_mode: str | None = None,
+    preferred_history_kind: str | None = None,
+    importance_threshold: int | None = None,
+    payload: dict | None = None,
+) -> dict:
+    repository = DecisionAssistantRepository()
+    normalized_profile_key = _normalize_profile_key(profile_key=profile_key, profile_label=profile_label)
+    if not normalized_profile_key:
+        return {
+            "status": "ignored",
+            "reason": "invalid_profile_key",
+            "instrument_id": instrument_id,
+            "profile_key": profile_key,
+        }
+    if repository.fetch_preference(instrument_id=instrument_id, profile_key=normalized_profile_key):
+        return {
+            "status": "ignored",
+            "reason": "profile_exists",
+            "instrument_id": instrument_id,
+            "profile_key": normalized_profile_key,
+        }
+    base = repository.fetch_preference(
+        instrument_id=instrument_id,
+        profile_key=source_profile_key or DecisionAssistantRepository.DEFAULT_PROFILE_KEY,
+    ) or _default_preference_payload(
+        instrument_id=instrument_id,
+        profile_key=source_profile_key or DecisionAssistantRepository.DEFAULT_PROFILE_KEY,
+    )
+    merged_payload = _normalize_preference_payload((base or {}).get("payload"))
+    merged_payload.update(dict(payload or {}))
+    merged_payload["profile_label"] = profile_label or merged_payload.get("profile_label") or normalized_profile_key
+    saved = repository.upsert_preference(
+        instrument_id=instrument_id,
+        profile_key=normalized_profile_key,
+        digest_mode=digest_mode or base.get("digest_mode") or "all_reminders",
+        preferred_history_kind=preferred_history_kind if preferred_history_kind is not None else base.get("preferred_history_kind"),
+        importance_threshold=int(importance_threshold or base.get("importance_threshold") or 70),
+        payload=merged_payload,
+    )
+    saved["status"] = "created"
+    saved["payload"] = _normalize_preference_payload(saved.get("payload"))
+    return saved
+
+
+def rename_operator_digest_preference_profile(
+    instrument_id: int,
+    *,
+    profile_key: str,
+    new_profile_key: str | None = None,
+    profile_label: str | None = None,
+) -> dict:
+    repository = DecisionAssistantRepository()
+    source = repository.fetch_preference(instrument_id=instrument_id, profile_key=profile_key)
+    if not source:
+        return {
+            "status": "ignored",
+            "reason": "profile_missing",
+            "instrument_id": instrument_id,
+            "profile_key": profile_key,
+        }
+    normalized_target_key = _normalize_profile_key(
+        profile_key=new_profile_key or profile_key,
+        profile_label=profile_label,
+    )
+    if not normalized_target_key:
+        return {
+            "status": "ignored",
+            "reason": "invalid_profile_key",
+            "instrument_id": instrument_id,
+            "profile_key": profile_key,
+        }
+    if (
+        profile_key == DecisionAssistantRepository.DEFAULT_PROFILE_KEY
+        and normalized_target_key != DecisionAssistantRepository.DEFAULT_PROFILE_KEY
+    ):
+        return {
+            "status": "ignored",
+            "reason": "default_profile_protected",
+            "instrument_id": instrument_id,
+            "profile_key": profile_key,
+        }
+    if normalized_target_key != profile_key and repository.fetch_preference(instrument_id=instrument_id, profile_key=normalized_target_key):
+        return {
+            "status": "ignored",
+            "reason": "profile_exists",
+            "instrument_id": instrument_id,
+            "profile_key": normalized_target_key,
+        }
+    merged_payload = _normalize_preference_payload(source.get("payload"))
+    if profile_label:
+        merged_payload["profile_label"] = profile_label
+    saved = repository.upsert_preference(
+        instrument_id=instrument_id,
+        profile_key=normalized_target_key,
+        digest_mode=source.get("digest_mode") or "all_reminders",
+        preferred_history_kind=source.get("preferred_history_kind"),
+        importance_threshold=int(source.get("importance_threshold") or 70),
+        payload=merged_payload,
+    )
+    if normalized_target_key != profile_key:
+        repository.delete_preference(instrument_id=instrument_id, profile_key=profile_key)
+    saved["status"] = "renamed"
+    saved["previous_profile_key"] = profile_key
+    saved["payload"] = _normalize_preference_payload(saved.get("payload"))
+    return saved
+
+
+def delete_operator_digest_preference_profile(
+    instrument_id: int,
+    *,
+    profile_key: str,
+) -> dict:
+    if profile_key == DecisionAssistantRepository.DEFAULT_PROFILE_KEY:
+        return {
+            "status": "ignored",
+            "reason": "default_profile_protected",
+            "instrument_id": instrument_id,
+            "profile_key": profile_key,
+        }
+    deleted = DecisionAssistantRepository().delete_preference(instrument_id=instrument_id, profile_key=profile_key)
+    return {
+        "status": "deleted" if deleted else "ignored",
+        "reason": None if deleted else "profile_missing",
+        "instrument_id": instrument_id,
+        "profile_key": profile_key,
+        "deleted_count": deleted,
+    }
+
+
 def save_operator_digest_preferences(
     instrument_id: int,
     *,
@@ -212,6 +357,7 @@ def get_operator_digest_history_archive(
     offset: int = 0,
     preferred_history_kind: str | None = None,
     digest_mode: str | None = None,
+    query_text: str | None = None,
     created_after=None,
     created_before=None,
 ) -> dict:
@@ -223,10 +369,13 @@ def get_operator_digest_history_archive(
         offset=offset,
         preferred_history_kind=preferred_history_kind,
         digest_mode=digest_mode,
+        query_text=query_text,
         created_after=created_after,
         created_before=created_before,
     )
     items = [_normalize_digest_history_row(row) for row in rows]
+    long_range_summary = build_digest_archive_long_range_summary(items)
+    baseline_summary = build_digest_archive_baseline_summary(items)
     return {
         "status": "active" if items else "idle",
         "instrument_id": instrument_id,
@@ -239,9 +388,12 @@ def get_operator_digest_history_archive(
         "filters": {
             "preferred_history_kind": preferred_history_kind,
             "digest_mode": digest_mode,
+            "query_text": query_text,
             "created_after": created_after,
             "created_before": created_before,
         },
+        "long_range_trend_summary": long_range_summary,
+        "baseline_pin_summary": baseline_summary,
     }
 
 
@@ -253,6 +405,7 @@ def compare_operator_digest_history_archive(
     right_id: int | None = None,
     preferred_history_kind: str | None = None,
     digest_mode: str | None = None,
+    query_text: str | None = None,
 ) -> dict:
     repository = DecisionAssistantRepository()
     if left_id and right_id:
@@ -268,11 +421,65 @@ def compare_operator_digest_history_archive(
             limit=2,
             preferred_history_kind=preferred_history_kind,
             digest_mode=digest_mode,
+            query_text=query_text,
         )
     items = [_normalize_digest_history_row(row) for row in rows]
     return _build_digest_archive_compare(items)
 
 
+def export_operator_digest_history_archive(
+    instrument_id: int,
+    *,
+    profile_key: str = DecisionAssistantRepository.DEFAULT_PROFILE_KEY,
+    limit: int = 20,
+    offset: int = 0,
+    preferred_history_kind: str | None = None,
+    digest_mode: str | None = None,
+    query_text: str | None = None,
+    created_after=None,
+    created_before=None,
+    left_id: int | None = None,
+    right_id: int | None = None,
+) -> dict:
+    archive = get_operator_digest_history_archive(
+        instrument_id=instrument_id,
+        profile_key=profile_key,
+        limit=limit,
+        offset=offset,
+        preferred_history_kind=preferred_history_kind,
+        digest_mode=digest_mode,
+        query_text=query_text,
+        created_after=created_after,
+        created_before=created_before,
+    )
+    compare = None
+    if left_id and right_id:
+        compare = compare_operator_digest_history_archive(
+            instrument_id=instrument_id,
+            profile_key=profile_key,
+            left_id=left_id,
+            right_id=right_id,
+            preferred_history_kind=preferred_history_kind,
+            digest_mode=digest_mode,
+            query_text=query_text,
+        )
+    timestamp = datetime.now(timezone.utc).strftime("%Y%m%dT%H%M%SZ")
+    return {
+        "status": archive.get("status") or "idle",
+        "export_kind": "digest_archive_bundle",
+        "generated_ts": datetime.now(timezone.utc).isoformat(),
+        "export_name": f"digest_archive_{profile_key}_{timestamp}.json",
+        "instrument_id": instrument_id,
+        "profile_key": profile_key,
+        "filters": dict(archive.get("filters") or {}),
+        "item_count": archive.get("item_count") or 0,
+        "items": list(archive.get("items") or []),
+        "long_range_trend_summary": dict(archive.get("long_range_trend_summary") or {}),
+        "baseline_pin_summary": dict(archive.get("baseline_pin_summary") or {}),
+        "compare": compare,
+    }
+
+
 def get_operator_workflow_archive(
     instrument_id: int,
     *,
@@ -282,6 +489,7 @@ def get_operator_workflow_archive(
     current_session: str | None = None,
     next_phase: str | None = None,
     next_artifact_kind: str | None = None,
+    query_text: str | None = None,
     created_after=None,
     created_before=None,
 ) -> dict:
@@ -294,10 +502,14 @@ def get_operator_workflow_archive(
         current_session=current_session,
         next_phase=next_phase,
         next_artifact_kind=next_artifact_kind,
+        query_text=query_text,
         created_after=created_after,
         created_before=created_before,
     )
     items = [_normalize_workflow_history_row(row) for row in rows]
+    operations_view = build_workflow_archive_operations_view(items)
+    long_range_summary = build_workflow_archive_long_range_summary(items)
+    baseline_summary = build_workflow_archive_baseline_summary(items)
     return {
         "status": "active" if items else "idle",
         "instrument_id": instrument_id,
@@ -311,9 +523,13 @@ def get_operator_workflow_archive(
             "current_session": current_session,
             "next_phase": next_phase,
             "next_artifact_kind": next_artifact_kind,
+            "query_text": query_text,
             "created_after": created_after,
             "created_before": created_before,
         },
+        "long_range_trend_summary": long_range_summary,
+        "baseline_pin_summary": baseline_summary,
+        **operations_view,
     }
 
 
@@ -325,6 +541,7 @@ def compare_operator_workflow_archive(
     right_id: int | None = None,
     current_session: str | None = None,
     next_phase: str | None = None,
+    query_text: str | None = None,
 ) -> dict:
     repository = DecisionAssistantRepository()
     if left_id and right_id:
@@ -340,11 +557,69 @@ def compare_operator_workflow_archive(
             limit=2,
             current_session=current_session,
             next_phase=next_phase,
+            query_text=query_text,
         )
     items = [_normalize_workflow_history_row(row) for row in rows]
     return _build_workflow_archive_compare(items)
 
 
+def export_operator_workflow_archive(
+    instrument_id: int,
+    *,
+    profile_key: str = DecisionAssistantRepository.DEFAULT_PROFILE_KEY,
+    limit: int = 20,
+    offset: int = 0,
+    current_session: str | None = None,
+    next_phase: str | None = None,
+    next_artifact_kind: str | None = None,
+    query_text: str | None = None,
+    created_after=None,
+    created_before=None,
+    left_id: int | None = None,
+    right_id: int | None = None,
+) -> dict:
+    archive = get_operator_workflow_archive(
+        instrument_id=instrument_id,
+        profile_key=profile_key,
+        limit=limit,
+        offset=offset,
+        current_session=current_session,
+        next_phase=next_phase,
+        next_artifact_kind=next_artifact_kind,
+        query_text=query_text,
+        created_after=created_after,
+        created_before=created_before,
+    )
+    compare = None
+    if left_id and right_id:
+        compare = compare_operator_workflow_archive(
+            instrument_id=instrument_id,
+            profile_key=profile_key,
+            left_id=left_id,
+            right_id=right_id,
+            current_session=current_session,
+            next_phase=next_phase,
+            query_text=query_text,
+        )
+    timestamp = datetime.now(timezone.utc).strftime("%Y%m%dT%H%M%SZ")
+    return {
+        "status": archive.get("status") or "idle",
+        "export_kind": "workflow_archive_bundle",
+        "generated_ts": datetime.now(timezone.utc).isoformat(),
+        "export_name": f"workflow_archive_{profile_key}_{timestamp}.json",
+        "instrument_id": instrument_id,
+        "profile_key": profile_key,
+        "filters": dict(archive.get("filters") or {}),
+        "item_count": archive.get("item_count") or 0,
+        "items": list(archive.get("items") or []),
+        "long_range_trend_summary": dict(archive.get("long_range_trend_summary") or {}),
+        "baseline_pin_summary": dict(archive.get("baseline_pin_summary") or {}),
+        "bridge_summary": dict(archive.get("bridge_summary") or {}),
+        "handoff_bundle_summary": _build_workflow_archive_handoff_bundle_summary(archive, compare),
+        "compare": compare,
+    }
+
+
 def get_alert_silent_window_status() -> dict:
     return AlertSilentWindowService().build_status()
 
@@ -372,6 +647,14 @@ def _normalize_preference_payload(payload) -> dict:
     return normalized
 
 
+def _normalize_profile_key(*, profile_key: str | None, profile_label: str | None = None) -> str | None:
+    seed = str(profile_key or profile_label or "").strip().lower()
+    normalized = re.sub(r"[^a-z0-9]+", "_", seed).strip("_")
+    if not normalized:
+        return None
+    return normalized[:64]
+
+
 def _default_preference_payload(*, instrument_id: int, profile_key: str) -> dict:
     return {
         "profile_key": profile_key,
@@ -506,3 +789,22 @@ def _build_workflow_archive_compare(items: list[dict]) -> dict:
         "count_delta": count_delta,
         "summary": f"Workflow archive compare shows {changed_count} changed phase brief(s) across the latest two snapshots.",
     }
+
+
+def _build_workflow_archive_handoff_bundle_summary(archive: dict, compare: dict | None) -> dict:
+    bridge_summary = dict(archive.get("bridge_summary") or {})
+    items = list(archive.get("items") or [])
+    latest = dict(items[0] or {})
+    compare_status = dict(compare or {}).get("status")
+    return {
+        "status": "active" if items else "idle",
+        "latest_snapshot_id": latest.get("snapshot_id"),
+        "latest_session": latest.get("current_session"),
+        "latest_next_phase": latest.get("next_phase"),
+        "latest_artifact_kind": latest.get("next_artifact_kind"),
+        "compare_status": compare_status or "idle",
+        "bridge_target_count": bridge_summary.get("bridge_target_count") or 0,
+        "next_handoff_prompt": bridge_summary.get("next_handoff_prompt"),
+        "next_replay_prompt": bridge_summary.get("next_replay_prompt"),
+        "next_note_prompt": bridge_summary.get("next_note_prompt"),
+    }

src/api/server.py

@@ -13,8 +13,12 @@ from src.api.harness import run_unattended_harness
 from src.api.live_operation_audits import get_recent_live_operation_audits
 from src.api.maintenance import get_maintenance_summary
 from src.api.alerts import (
+    create_operator_digest_preference_profile,
     compare_operator_digest_history_archive,
     compare_operator_workflow_archive,
+    delete_operator_digest_preference_profile,
+    export_operator_digest_history_archive,
+    export_operator_workflow_archive,
     get_alert_deliveries,
     get_operator_digest_history_archive,
     get_alert_quality_summary,
@@ -24,6 +28,7 @@ from src.api.alerts import (
     get_operator_focus_state,
     get_operator_workflow_archive,
     list_operator_digest_preference_profiles,
+    rename_operator_digest_preference_profile,
     save_operator_focus_state,
     save_operator_digest_preferences,
 )
@@ -135,11 +140,16 @@ def build_api_dependencies() -> dict:
         "trading_alert_operator_digest": get_operator_digest,
         "trading_alert_operator_digest_preferences": get_operator_digest_preferences,
         "trading_alert_operator_digest_profiles": list_operator_digest_preference_profiles,
+        "trading_alert_operator_digest_profiles_create": create_operator_digest_preference_profile,
+        "trading_alert_operator_digest_profiles_rename": rename_operator_digest_preference_profile,
+        "trading_alert_operator_digest_profiles_delete": delete_operator_digest_preference_profile,
         "trading_alert_operator_digest_preferences_save": save_operator_digest_preferences,
         "trading_alert_operator_digest_history_archive": get_operator_digest_history_archive,
         "trading_alert_operator_digest_history_compare": compare_operator_digest_history_archive,
+        "trading_alert_operator_digest_history_export": export_operator_digest_history_archive,
         "trading_workflow_archive": get_operator_workflow_archive,
         "trading_workflow_archive_compare": compare_operator_workflow_archive,
+        "trading_workflow_archive_export": export_operator_workflow_archive,
         "trading_assistant_focus": get_operator_focus_state,
         "trading_assistant_focus_save": save_operator_focus_state,
         "trading_alert_silent_window_status": get_alert_silent_window_status,
@@ -334,6 +344,42 @@ def route_request(
     if method == "GET" and path == "/trading/alerts/digest-preferences/profiles":
         instrument_id = _require_int(query, "instrument_id")
         return 200, deps["trading_alert_operator_digest_profiles"](instrument_id=instrument_id)
+    if method == "POST" and path == "/trading/alerts/digest-preferences/profiles":
+        instrument_id = _require_json_int(body or {}, "instrument_id")
+        return 200, deps["trading_alert_operator_digest_profiles_create"](
+            instrument_id=instrument_id,
+            profile_key=_get_body_value(body, "profile_key"),
+            profile_label=_get_body_value(body, "profile_label"),
+            source_profile_key=_get_body_value(body, "source_profile_key"),
+            digest_mode=_get_body_value(body, "digest_mode"),
+            preferred_history_kind=_get_body_value(body, "preferred_history_kind"),
+            importance_threshold=_get_body_optional_int(body, "importance_threshold"),
+            payload=body.get("payload") if isinstance(body, dict) else None,
+        )
+    if (
+        method == "POST"
+        and len(path_parts) == 6
+        and path_parts[:4] == ["trading", "alerts", "digest-preferences", "profiles"]
+        and path_parts[5] == "rename"
+    ):
+        instrument_id = _require_json_int(body or {}, "instrument_id")
+        return 200, deps["trading_alert_operator_digest_profiles_rename"](
+            instrument_id=instrument_id,
+            profile_key=path_parts[4],
+            new_profile_key=_get_body_value(body, "new_profile_key"),
+            profile_label=_get_body_value(body, "profile_label"),
+        )
+    if (
+        method == "POST"
+        and len(path_parts) == 6
+        and path_parts[:4] == ["trading", "alerts", "digest-preferences", "profiles"]
+        and path_parts[5] == "delete"
+    ):
+        instrument_id = _require_json_int(body or {}, "instrument_id")
+        return 200, deps["trading_alert_operator_digest_profiles_delete"](
+            instrument_id=instrument_id,
+            profile_key=path_parts[4],
+        )
     if method == "POST" and path == "/trading/alerts/digest-preferences":
         instrument_id = _require_json_int(body or {}, "instrument_id")
         return 200, deps["trading_alert_operator_digest_preferences_save"](
@@ -353,9 +399,25 @@ def route_request(
             offset=_get_int(query, "offset", 0),
             preferred_history_kind=_get_single(query, "preferred_history_kind"),
             digest_mode=_get_single(query, "digest_mode"),
+            query_text=_get_single(query, "query_text"),
             created_after=_get_single(query, "created_after"),
             created_before=_get_single(query, "created_before"),
         )
+    if method == "GET" and path == "/trading/alerts/digest-history-archive/export":
+        instrument_id = _require_int(query, "instrument_id")
+        return 200, deps["trading_alert_operator_digest_history_export"](
+            instrument_id=instrument_id,
+            profile_key=_get_single(query, "profile_key", "local_operator"),
+            limit=_get_int(query, "limit", 20),
+            offset=_get_int(query, "offset", 0),
+            preferred_history_kind=_get_single(query, "preferred_history_kind"),
+            digest_mode=_get_single(query, "digest_mode"),
+            query_text=_get_single(query, "query_text"),
+            created_after=_get_single(query, "created_after"),
+            created_before=_get_single(query, "created_before"),
+            left_id=_get_optional_int(query, "left_id"),
+            right_id=_get_optional_int(query, "right_id"),
+        )
     if method == "GET" and path == "/trading/alerts/digest-history-archive/compare":
         instrument_id = _require_int(query, "instrument_id")
         return 200, deps["trading_alert_operator_digest_history_compare"](
@@ -365,6 +427,7 @@ def route_request(
             right_id=_get_optional_int(query, "right_id"),
             preferred_history_kind=_get_single(query, "preferred_history_kind"),
             digest_mode=_get_single(query, "digest_mode"),
+            query_text=_get_single(query, "query_text"),
         )
     if method == "GET" and path == "/trading/workflow-archive":
         instrument_id = _require_int(query, "instrument_id")
@@ -376,9 +439,26 @@ def route_request(
             current_session=_get_single(query, "current_session"),
             next_phase=_get_single(query, "next_phase"),
             next_artifact_kind=_get_single(query, "next_artifact_kind"),
+            query_text=_get_single(query, "query_text"),
             created_after=_get_single(query, "created_after"),
             created_before=_get_single(query, "created_before"),
         )
+    if method == "GET" and path == "/trading/workflow-archive/export":
+        instrument_id = _require_int(query, "instrument_id")
+        return 200, deps["trading_workflow_archive_export"](
+            instrument_id=instrument_id,
+            profile_key=_get_single(query, "profile_key", "local_operator"),
+            limit=_get_int(query, "limit", 20),
+            offset=_get_int(query, "offset", 0),
+            current_session=_get_single(query, "current_session"),
+            next_phase=_get_single(query, "next_phase"),
+            next_artifact_kind=_get_single(query, "next_artifact_kind"),
+            query_text=_get_single(query, "query_text"),
+            created_after=_get_single(query, "created_after"),
+            created_before=_get_single(query, "created_before"),
+            left_id=_get_optional_int(query, "left_id"),
+            right_id=_get_optional_int(query, "right_id"),
+        )
     if method == "GET" and path == "/trading/workflow-archive/compare":
         instrument_id = _require_int(query, "instrument_id")
         return 200, deps["trading_workflow_archive_compare"](
@@ -388,6 +468,7 @@ def route_request(
             right_id=_get_optional_int(query, "right_id"),
             current_session=_get_single(query, "current_session"),
             next_phase=_get_single(query, "next_phase"),
+            query_text=_get_single(query, "query_text"),
         )
     if method == "GET" and path == "/trading/assistant-focus":
         instrument_id = _require_int(query, "instrument_id")
@@ -453,7 +534,10 @@ def route_request(
         )
     if method == "GET" and path == "/trading/review-packet":
         instrument_id = _require_int(query, "instrument_id")
-        return 200, deps["trading_review_packet"](instrument_id=instrument_id)
+        return 200, deps["trading_review_packet"](
+            instrument_id=instrument_id,
+            profile_key=_get_single(query, "profile_key", "local_operator"),
+        )
     if method == "GET" and path == "/trading/research-council/latest":
         instrument_id = _require_int(query, "instrument_id")
         payload = deps["trading_research_council_latest"](instrument_id=instrument_id)

src/api/trading_runtime.py

@@ -456,8 +456,12 @@ def create_trading_operator_note(
     )
 
 
-def get_trading_review_packet(instrument_id: int) -> dict:
-    return ReviewPacketService().build_packet(instrument_id=instrument_id)
+def get_trading_review_packet(
+    instrument_id: int,
+    *,
+    profile_key: str = "local_operator",
+) -> dict:
+    return ReviewPacketService().build_packet(instrument_id=instrument_id, profile_key=profile_key)
 
 
 def build_live_broker_service() -> LiveBrokerService:

src/repositories/decision_assistant_repository.py

@@ -100,6 +100,24 @@ class DecisionAssistantRepository:
             ).mappings().one()
         return dict(row)
 
+    def delete_preference(self, *, instrument_id: int, profile_key: str = DEFAULT_PROFILE_KEY) -> int:
+        sql = text(
+            """
+            delete from decision_assistant_preferences
+            where instrument_id = :instrument_id
+              and profile_key = :profile_key
+            """
+        )
+        with SessionLocal.begin() as session:
+            result = session.execute(
+                sql,
+                {
+                    "instrument_id": instrument_id,
+                    "profile_key": profile_key,
+                },
+            )
+        return int(result.rowcount or 0)
+
     def fetch_recent_digest_history(
         self,
         *,
@@ -109,6 +127,7 @@ class DecisionAssistantRepository:
         offset: int = 0,
         digest_mode: str | None = None,
         preferred_history_kind: str | None = None,
+        query_text: str | None = None,
         created_after=None,
         created_before=None,
     ) -> list[dict]:
@@ -128,6 +147,16 @@ class DecisionAssistantRepository:
         if preferred_history_kind:
             filters.append("preferred_history_kind = :preferred_history_kind")
             params["preferred_history_kind"] = preferred_history_kind
+        if query_text:
+            filters.append(
+                """
+                (
+                    coalesce(top_summary, '') ilike :query_text_like
+                    or cast(payload as text) ilike :query_text_like
+                )
+                """
+            )
+            params["query_text_like"] = f"%{query_text.strip()}%"
         if created_after is not None:
             filters.append("created_ts >= :created_after")
             params["created_after"] = created_after
@@ -244,6 +273,7 @@ class DecisionAssistantRepository:
         current_session: str | None = None,
         next_phase: str | None = None,
         next_artifact_kind: str | None = None,
+        query_text: str | None = None,
         created_after=None,
         created_before=None,
     ) -> list[dict]:
@@ -266,6 +296,18 @@ class DecisionAssistantRepository:
         if next_artifact_kind:
             filters.append("next_artifact_kind = :next_artifact_kind")
             params["next_artifact_kind"] = next_artifact_kind
+        if query_text:
+            filters.append(
+                """
+                (
+                    coalesce(current_session, '') ilike :query_text_like
+                    or coalesce(next_phase, '') ilike :query_text_like
+                    or coalesce(next_artifact_kind, '') ilike :query_text_like
+                    or cast(payload as text) ilike :query_text_like
+                )
+                """
+            )
+            params["query_text_like"] = f"%{query_text.strip()}%"
         if created_after is not None:
             filters.append("created_ts >= :created_after")
             params["created_after"] = created_after

src/services/market_state/exchange_status_provider.py

@@ -264,7 +264,7 @@ def _normalize_announcement_item(row: dict) -> dict:
 
 
 def _fetch_okx_json(*, url: str, timeout: int, proxy_url: str, headers: dict, payload_error: str) -> dict:
-    attempt_timeout = max(1, min(int(timeout or 1), 4))
+    attempt_timeout = max(1, min(int(timeout or 1), 2))
     attempts = []
     if proxy_url:
         attempts.append({"timeout": attempt_timeout, "follow_redirects": True, "proxy": proxy_url, "trust_env": False})

src/services/market_state/external_chart_provider.py

@@ -100,7 +100,7 @@ class BinanceChartReferenceProvider:
         return str((instrument or {}).get("symbol") or "").strip() or None
 
     def _fetch_payload(self, *, api_symbol: str, interval: str) -> list:
-        attempt_timeout = max(1, min(int(self.timeout or 1), 4))
+        attempt_timeout = max(1, min(int(self.timeout or 1), 2))
         attempts = []
         if self.proxy_url:
             attempts.append({"timeout": attempt_timeout, "follow_redirects": True, "proxy": self.proxy_url, "trust_env": False})

src/services/market_state/market_state_service.py

@@ -155,14 +155,17 @@ class MarketStateService:
         max_candle_age_seconds: int = DEFAULT_MAX_CANDLE_AGE_SECONDS,
         news_windows: Optional[list[dict]] = None,
         opening_volatility_minutes: int = 0,
+        cached_news_only: bool = False,
     ) -> MarketState:
         observed_ts = _ensure_aware(now or datetime.now(timezone.utc))
         runtime_state = None
         if self.market_truth_supervisor is not None:
-            runtime_state = self.market_truth_supervisor.current_state(
+            runtime_state = _current_runtime_state(
+                supervisor=self.market_truth_supervisor,
                 instrument_id=instrument_id,
                 timeframe=timeframe,
                 now=observed_ts,
+                cached_news_only=cached_news_only,
             )
 
         latest_candle = self.repository.fetch_latest_candle(instrument_id=instrument_id, timeframe=timeframe)
@@ -441,3 +444,19 @@ def _timeframe_to_seconds(timeframe: str) -> int:
     if text.endswith("d"):
         return int(text[:-1]) * 24 * 60 * 60
     raise ValueError(f"unsupported_timeframe:{timeframe}")
+
+
+def _current_runtime_state(*, supervisor, instrument_id: int, timeframe: str, now: datetime, cached_news_only: bool):
+    try:
+        return supervisor.current_state(
+            instrument_id=instrument_id,
+            timeframe=timeframe,
+            now=now,
+            cached_news_only=cached_news_only,
+        )
+    except TypeError:
+        return supervisor.current_state(
+            instrument_id=instrument_id,
+            timeframe=timeframe,
+            now=now,
+        )

src/services/market_state/market_truth_supervisor.py

@@ -108,7 +108,14 @@ class MarketTruthSupervisor:
         if self.enabled:
             self.start()
 
-    def current_state(self, *, instrument_id: int, timeframe: str, now: Optional[datetime] = None) -> MarketTruthRuntimeState:
+    def current_state(
+        self,
+        *,
+        instrument_id: int,
+        timeframe: str,
+        now: Optional[datetime] = None,
+        cached_news_only: bool = False,
+    ) -> MarketTruthRuntimeState:
         now = _ensure_aware(now or datetime.now(timezone.utc))
         self.ensure_subscription(instrument_id=instrument_id, timeframe=timeframe)
         latest_candle = self.repository.fetch_latest_candle(instrument_id=instrument_id, timeframe=timeframe)
@@ -126,7 +133,11 @@ class MarketTruthSupervisor:
             now=now,
             gap_multiplier=self.gap_multiplier,
         )
-        news_snapshot = self._get_news_snapshot(now=now, force_refresh=True)
+        news_snapshot = self._get_news_snapshot(
+            now=now,
+            force_refresh=not cached_news_only,
+            cached_only=cached_news_only,
+        )
         if self._should_fallback(stream_state=stream_state, data_gap_state=data_gap_state) and self._fallback_allowed(
             instrument_id=instrument_id,
             timeframe=timeframe,
@@ -526,9 +537,21 @@ class MarketTruthSupervisor:
             return True
         return (now - last_attempt).total_seconds() >= max(self.fallback_cooldown_seconds, 1)
 
-    def _get_news_snapshot(self, *, now: datetime, force_refresh: bool = False) -> NewsProviderSnapshot:
+    def _get_news_snapshot(
+        self,
+        *,
+        now: datetime,
+        force_refresh: bool = False,
+        cached_only: bool = False,
+    ) -> NewsProviderSnapshot:
         if not settings.news_provider_enabled:
             return self.news_provider.evaluate_cached_snapshot(snapshot=None, now=now)
+        if cached_only:
+            snapshot = self._news_snapshot
+            if snapshot is None:
+                snapshot = build_snapshot_from_row(self.repository.fetch_latest_news_provider_snapshot())
+                self._news_snapshot = snapshot
+            return self.news_provider.evaluate_cached_snapshot(snapshot=snapshot, now=now)
         refresh_interval = max(settings.news_provider_refresh_seconds, 1)
         should_refresh = force_refresh or self._news_snapshot is None or self._last_news_refresh_ts is None
         if not should_refresh and self._last_news_refresh_ts is not None:

src/services/market_state/news_event_provider.py

@@ -118,7 +118,8 @@ class BaseNewsProvider:
         return window_start <= now <= window_end
 
     def _client_kwargs(self) -> dict:
-        kwargs = {"timeout": self.timeout, "follow_redirects": True}
+        attempt_timeout = max(1, min(int(self.timeout or 1), 2))
+        kwargs = {"timeout": attempt_timeout, "follow_redirects": True}
         if self.proxy_url:
             kwargs["proxy"] = self.proxy_url
             kwargs["trust_env"] = False

src/services/operator/archive_long_range_projection.py

@@ -0,0 +1,179 @@
+from __future__ import annotations
+
+from datetime import datetime
+
+
+def build_digest_archive_long_range_summary(items: list[dict]) -> dict:
+    if not items:
+        return {
+            "status": "idle",
+            "window_basis": "query_result_window",
+            "snapshot_count": 0,
+            "recent_7d_count": 0,
+            "recent_30d_count": 0,
+            "covered_day_span": 0,
+            "trend_status": "no_history",
+            "next_review_prompt": None,
+        }
+    latest_dt = _safe_dt(items[0].get("created_ts"))
+    oldest_dt = _safe_dt(items[-1].get("created_ts"))
+    recent_7d = [item for item in items if _within_days(item.get("created_ts"), latest_dt, 7)]
+    recent_30d = [item for item in items if _within_days(item.get("created_ts"), latest_dt, 30)]
+    high_value_7d = sum(int(item.get("high_value_reminder_count") or 0) for item in recent_7d)
+    high_value_30d = sum(int(item.get("high_value_reminder_count") or 0) for item in recent_30d)
+    covered_day_span = _covered_day_span(latest_dt, oldest_dt)
+    trend_status = _archive_trend_status(
+        snapshot_count=len(items),
+        recent_7d_count=len(recent_7d),
+        recent_30d_count=len(recent_30d),
+        covered_day_span=covered_day_span,
+    )
+    return {
+        "status": "active",
+        "window_basis": "query_result_window",
+        "snapshot_count": len(items),
+        "latest_created_ts": items[0].get("created_ts"),
+        "oldest_created_ts": items[-1].get("created_ts"),
+        "covered_day_span": covered_day_span,
+        "recent_7d_count": len(recent_7d),
+        "recent_30d_count": len(recent_30d),
+        "recent_7d_high_value_count": high_value_7d,
+        "recent_30d_high_value_count": high_value_30d,
+        "trend_status": trend_status,
+        "next_review_prompt": f"Review 7d vs 30d digest trend before rotating the next archive baseline.",
+    }
+
+
+def build_digest_archive_baseline_summary(items: list[dict]) -> dict:
+    if not items:
+        return {
+            "status": "idle",
+            "baseline_snapshot_id": None,
+            "latest_snapshot_id": None,
+            "compare_ready": False,
+            "suggested_compare_prompt": None,
+        }
+    latest = dict(items[0] or {})
+    baseline = dict(items[-1] or {})
+    return {
+        "status": "active" if len(items) >= 2 else "insufficient_history",
+        "latest_snapshot_id": latest.get("snapshot_id"),
+        "latest_created_ts": latest.get("created_ts"),
+        "latest_summary": latest.get("top_summary"),
+        "baseline_snapshot_id": baseline.get("snapshot_id"),
+        "baseline_created_ts": baseline.get("created_ts"),
+        "baseline_summary": baseline.get("top_summary"),
+        "compare_ready": len(items) >= 2,
+        "suggested_compare_prompt": (
+            f"Pin digest snapshot {baseline.get('snapshot_id')} as the baseline and compare it against {latest.get('snapshot_id')}."
+            if len(items) >= 2
+            else "Need at least two digest archive snapshots before pinning a baseline."
+        ),
+    }
+
+
+def build_workflow_archive_long_range_summary(items: list[dict]) -> dict:
+    if not items:
+        return {
+            "status": "idle",
+            "window_basis": "query_result_window",
+            "snapshot_count": 0,
+            "recent_7d_count": 0,
+            "recent_30d_count": 0,
+            "covered_day_span": 0,
+            "trend_status": "no_history",
+            "next_review_prompt": None,
+        }
+    latest_dt = _safe_dt(items[0].get("created_ts"))
+    oldest_dt = _safe_dt(items[-1].get("created_ts"))
+    recent_7d = [item for item in items if _within_days(item.get("created_ts"), latest_dt, 7)]
+    recent_30d = [item for item in items if _within_days(item.get("created_ts"), latest_dt, 30)]
+    covered_day_span = _covered_day_span(latest_dt, oldest_dt)
+    distinct_sessions = len({str(item.get("current_session") or "unknown") for item in items})
+    trend_status = _archive_trend_status(
+        snapshot_count=len(items),
+        recent_7d_count=len(recent_7d),
+        recent_30d_count=len(recent_30d),
+        covered_day_span=covered_day_span,
+    )
+    return {
+        "status": "active",
+        "window_basis": "query_result_window",
+        "snapshot_count": len(items),
+        "latest_created_ts": items[0].get("created_ts"),
+        "oldest_created_ts": items[-1].get("created_ts"),
+        "covered_day_span": covered_day_span,
+        "recent_7d_count": len(recent_7d),
+        "recent_30d_count": len(recent_30d),
+        "recent_7d_ready_count": sum(int(item.get("ready_for_handoff_count") or 0) for item in recent_7d),
+        "recent_30d_ready_count": sum(int(item.get("ready_for_handoff_count") or 0) for item in recent_30d),
+        "distinct_session_count": distinct_sessions,
+        "trend_status": trend_status,
+        "next_review_prompt": "Review 7d vs 30d workflow archive trend before pinning the next baseline.",
+    }
+
+
+def build_workflow_archive_baseline_summary(items: list[dict]) -> dict:
+    if not items:
+        return {
+            "status": "idle",
+            "baseline_snapshot_id": None,
+            "latest_snapshot_id": None,
+            "compare_ready": False,
+            "suggested_compare_prompt": None,
+        }
+    latest = dict(items[0] or {})
+    baseline = dict(items[-1] or {})
+    return {
+        "status": "active" if len(items) >= 2 else "insufficient_history",
+        "latest_snapshot_id": latest.get("snapshot_id"),
+        "latest_created_ts": latest.get("created_ts"),
+        "latest_next_phase": latest.get("next_phase"),
+        "latest_session": latest.get("current_session"),
+        "baseline_snapshot_id": baseline.get("snapshot_id"),
+        "baseline_created_ts": baseline.get("created_ts"),
+        "baseline_next_phase": baseline.get("next_phase"),
+        "baseline_session": baseline.get("current_session"),
+        "compare_ready": len(items) >= 2,
+        "suggested_compare_prompt": (
+            f"Pin workflow snapshot {baseline.get('snapshot_id')} as the baseline and compare it against {latest.get('snapshot_id')}."
+            if len(items) >= 2
+            else "Need at least two workflow archive snapshots before pinning a baseline."
+        ),
+    }
+
+
+def _safe_dt(value):
+    if not value:
+        return None
+    text = str(value)
+    try:
+        return datetime.fromisoformat(text.replace("Z", "+00:00"))
+    except Exception:
+        return None
+
+
+def _within_days(value, latest_dt, days: int) -> bool:
+    current_dt = _safe_dt(value)
+    if not latest_dt or not current_dt:
+        return False
+    delta_days = (latest_dt - current_dt).total_seconds() / 86400
+    return delta_days <= float(days)
+
+
+def _covered_day_span(latest_dt, oldest_dt) -> int:
+    if not latest_dt or not oldest_dt:
+        return 0
+    return max(0, int(round((latest_dt - oldest_dt).total_seconds() / 86400)))
+
+
+def _archive_trend_status(*, snapshot_count: int, recent_7d_count: int, recent_30d_count: int, covered_day_span: int) -> str:
+    if snapshot_count <= 1:
+        return "thin_history"
+    if covered_day_span >= 30 and recent_30d_count >= 6:
+        return "long_range_ready"
+    if covered_day_span >= 7 and recent_7d_count >= 3:
+        return "weekly_ready"
+    if recent_30d_count > 0:
+        return "recent_window_only"
+    return "thin_history"

src/services/operator/workflow_archive_projection.py

@@ -0,0 +1,276 @@
+from __future__ import annotations
+
+from datetime import datetime
+
+
+def build_workflow_archive_operations_view(items: list[dict]) -> dict:
+    daily_groups = _build_workflow_archive_daily_groups(items)
+    session_groups = _build_workflow_archive_session_groups(items)
+    operations_summary = _build_workflow_archive_operations_summary(
+        items=items,
+        daily_groups=daily_groups,
+        session_groups=session_groups,
+    )
+    multi_day_drift_review = _build_workflow_archive_multi_day_drift_review(daily_groups)
+    bridge_summary = _build_workflow_archive_bridge_summary(items)
+    return {
+        "daily_groups": daily_groups,
+        "daily_group_count": len(daily_groups),
+        "session_groups": session_groups,
+        "session_group_count": len(session_groups),
+        "operations_summary": operations_summary,
+        "multi_day_drift_review": multi_day_drift_review,
+        "bridge_summary": bridge_summary,
+    }
+
+
+def _build_workflow_archive_daily_groups(items: list[dict]) -> list[dict]:
+    buckets: dict[str, list[dict]] = {}
+    for item in items:
+        group_date = _workflow_archive_group_date(item.get("created_ts"))
+        buckets.setdefault(group_date, []).append(dict(item))
+    groups = []
+    for group_date, rows in buckets.items():
+        ordered = sorted(rows, key=lambda row: str(row.get("created_ts") or ""), reverse=True)
+        representative = dict(ordered[0] or {})
+        sessions = sorted({str(row.get("current_session") or "unknown") for row in ordered})
+        next_phase_counts = _count_values([row.get("next_phase") for row in ordered])
+        phase_names = _workflow_archive_phase_names(ordered)
+        groups.append(
+            {
+                "group_date": group_date,
+                "status": "active",
+                "snapshot_count": len(ordered),
+                "latest_created_ts": representative.get("created_ts"),
+                "current_session": representative.get("current_session"),
+                "sessions": sessions,
+                "session_count": len(sessions),
+                "next_phase": representative.get("next_phase"),
+                "next_phase_counts": next_phase_counts,
+                "next_artifact_kind": representative.get("next_artifact_kind"),
+                "artifact_count": int(representative.get("artifact_count") or 0),
+                "ready_for_handoff_count": int(representative.get("ready_for_handoff_count") or 0),
+                "needs_update_count": int(representative.get("needs_update_count") or 0),
+                "pending_capture_count": int(representative.get("pending_capture_count") or 0),
+                "phase_names": phase_names,
+                "phase_name_count": len(phase_names),
+                "workflow_body": representative.get("workflow_body"),
+                "next_artifact_prompt": representative.get("next_artifact_prompt"),
+                "representative_snapshot_id": representative.get("snapshot_id"),
+                "representative_phase_briefs": list(representative.get("phase_briefs") or []),
+            }
+        )
+    groups.sort(key=lambda row: str(row.get("group_date") or ""), reverse=True)
+    return groups[:14]
+
+
+def _build_workflow_archive_session_groups(items: list[dict]) -> list[dict]:
+    buckets: dict[str, list[dict]] = {}
+    for item in items:
+        session_code = str(item.get("current_session") or "unknown")
+        buckets.setdefault(session_code, []).append(dict(item))
+    groups = []
+    for session_code, rows in buckets.items():
+        ordered = sorted(rows, key=lambda row: str(row.get("created_ts") or ""), reverse=True)
+        representative = dict(ordered[0] or {})
+        day_count = len({str(_workflow_archive_group_date(row.get("created_ts"))) for row in ordered})
+        phase_names = _workflow_archive_phase_names(ordered)
+        groups.append(
+            {
+                "session_code": session_code,
+                "status": "active",
+                "snapshot_count": len(ordered),
+                "day_count": day_count,
+                "latest_created_ts": representative.get("created_ts"),
+                "latest_group_date": _workflow_archive_group_date(representative.get("created_ts")),
+                "latest_next_phase": representative.get("next_phase"),
+                "artifact_count": int(representative.get("artifact_count") or 0),
+                "ready_for_handoff_count": int(representative.get("ready_for_handoff_count") or 0),
+                "needs_update_count": int(representative.get("needs_update_count") or 0),
+                "pending_capture_count": int(representative.get("pending_capture_count") or 0),
+                "phase_names": phase_names,
+                "phase_name_count": len(phase_names),
+            }
+        )
+    groups.sort(
+        key=lambda row: (
+            int(row.get("snapshot_count") or 0),
+            str(row.get("latest_created_ts") or ""),
+            str(row.get("session_code") or ""),
+        ),
+        reverse=True,
+    )
+    return groups[:10]
+
+
+def _build_workflow_archive_operations_summary(*, items: list[dict], daily_groups: list[dict], session_groups: list[dict]) -> dict:
+    if not items:
+        return {
+            "status": "idle",
+            "snapshot_count": 0,
+            "day_group_count": 0,
+            "session_group_count": 0,
+            "latest_group_date": None,
+            "most_active_day": None,
+            "most_active_session": None,
+            "next_operations_prompt": None,
+        }
+    latest_day = dict(daily_groups[0] or {})
+    top_session = dict(session_groups[0] or {})
+    return {
+        "status": "active",
+        "snapshot_count": len(items),
+        "day_group_count": len(daily_groups),
+        "session_group_count": len(session_groups),
+        "latest_group_date": latest_day.get("group_date"),
+        "latest_snapshot_count": latest_day.get("snapshot_count"),
+        "most_active_day": latest_day.get("group_date"),
+        "most_active_day_session_count": latest_day.get("session_count"),
+        "most_active_session": top_session.get("session_code"),
+        "most_active_session_snapshot_count": top_session.get("snapshot_count"),
+        "phase_name_count": len(_workflow_archive_phase_names(items)),
+        "next_operations_prompt": (
+            f"Start with {latest_day.get('group_date')} archive group, then compare it against the previous day for workflow drift."
+            if latest_day.get("group_date")
+            else None
+        ),
+    }
+
+
+def _build_workflow_archive_multi_day_drift_review(daily_groups: list[dict]) -> dict:
+    if len(daily_groups) < 2:
+        return {
+            "status": "insufficient_history",
+            "left_group_date": None,
+            "right_group_date": None,
+            "phase_changes": [],
+            "summary": "Need at least two workflow day groups to review multi-day drift.",
+        }
+    left = dict(daily_groups[0] or {})
+    right = dict(daily_groups[1] or {})
+    left_map = {
+        str(item.get("phase") or "workflow"): str(item.get("archive_body") or item.get("detail") or "")
+        for item in list(left.get("representative_phase_briefs") or [])
+    }
+    right_map = {
+        str(item.get("phase") or "workflow"): str(item.get("archive_body") or item.get("detail") or "")
+        for item in list(right.get("representative_phase_briefs") or [])
+    }
+    phase_changes = []
+    for phase in sorted(set(left_map) | set(right_map)):
+        phase_changes.append(
+            {
+                "phase": phase,
+                "changed": str(left_map.get(phase) or "") != str(right_map.get(phase) or ""),
+                "left_body": left_map.get(phase),
+                "right_body": right_map.get(phase),
+            }
+        )
+    changed_phase_count = sum(1 for item in phase_changes if item.get("changed"))
+    session_changed = str(left.get("current_session") or "") != str(right.get("current_session") or "")
+    next_phase_changed = str(left.get("next_phase") or "") != str(right.get("next_phase") or "")
+    drift_status = "drift_detected" if changed_phase_count or session_changed or next_phase_changed else "stable_repeat"
+    return {
+        "status": "active",
+        "drift_status": drift_status,
+        "left_group_date": left.get("group_date"),
+        "right_group_date": right.get("group_date"),
+        "changed_phase_count": changed_phase_count,
+        "session_changed": session_changed,
+        "next_phase_changed": next_phase_changed,
+        "count_delta": {
+            "artifact_count": int(left.get("artifact_count") or 0) - int(right.get("artifact_count") or 0),
+            "ready_for_handoff_count": int(left.get("ready_for_handoff_count") or 0) - int(right.get("ready_for_handoff_count") or 0),
+            "needs_update_count": int(left.get("needs_update_count") or 0) - int(right.get("needs_update_count") or 0),
+            "pending_capture_count": int(left.get("pending_capture_count") or 0) - int(right.get("pending_capture_count") or 0),
+            "snapshot_count": int(left.get("snapshot_count") or 0) - int(right.get("snapshot_count") or 0),
+        },
+        "phase_changes": phase_changes,
+        "summary": (
+            f"Workflow drift review compares {left.get('group_date')} vs {right.get('group_date')}: "
+            f"{changed_phase_count} changed phase brief(s), session_changed={str(session_changed).lower()}, "
+            f"next_phase_changed={str(next_phase_changed).lower()}."
+        ),
+        "next_drift_prompt": (
+            f"Review {left.get('group_date')} against {right.get('group_date')} before promoting the next workflow cadence."
+        ),
+    }
+
+
+def _build_workflow_archive_bridge_summary(items: list[dict]) -> dict:
+    if not items:
+        return {
+            "status": "idle",
+            "bridge_target_count": 0,
+            "replay_ready_count": 0,
+            "note_ready_count": 0,
+            "handoff_ready_count": 0,
+            "latest_snapshot_id": None,
+            "latest_session": None,
+            "latest_next_phase": None,
+            "next_replay_prompt": None,
+            "next_note_prompt": None,
+            "next_handoff_prompt": None,
+        }
+    latest = dict(items[0] or {})
+    replay_ready_count = sum(1 for item in items if item.get("workflow_body") or item.get("phase_briefs"))
+    note_ready_count = sum(1 for item in items if item.get("workflow_body") or item.get("next_artifact_prompt"))
+    handoff_ready_count = sum(1 for item in items if item.get("next_phase") or item.get("next_artifact_kind"))
+    return {
+        "status": "active",
+        "bridge_target_count": 3,
+        "replay_ready_count": replay_ready_count,
+        "note_ready_count": note_ready_count,
+        "handoff_ready_count": handoff_ready_count,
+        "latest_snapshot_id": latest.get("snapshot_id"),
+        "latest_session": latest.get("current_session"),
+        "latest_next_phase": latest.get("next_phase"),
+        "latest_artifact_kind": latest.get("next_artifact_kind"),
+        "next_replay_prompt": (
+            f"Open replay with workflow snapshot {latest.get('snapshot_id')} from {latest.get('current_session') or 'unknown session'}."
+            if latest.get("snapshot_id")
+            else None
+        ),
+        "next_note_prompt": (
+            f"Prefill a review note from workflow snapshot {latest.get('snapshot_id')} before handoff."
+            if latest.get("snapshot_id")
+            else None
+        ),
+        "next_handoff_prompt": (
+            f"Bridge workflow snapshot {latest.get('snapshot_id')} into the next {latest.get('next_phase') or 'workflow'} handoff."
+            if latest.get("snapshot_id")
+            else None
+        ),
+    }
+
+
+def _workflow_archive_group_date(value) -> str:
+    if not value:
+        return "unknown_date"
+    text = str(value)
+    try:
+        parsed = datetime.fromisoformat(text.replace("Z", "+00:00"))
+        return parsed.date().isoformat()
+    except Exception:
+        return text[:10]
+
+
+def _workflow_archive_phase_names(items: list[dict]) -> list[str]:
+    phase_names = []
+    seen = set()
+    for item in items:
+        for phase in list(item.get("phase_briefs") or []):
+            name = str(phase.get("phase") or "")
+            if not name or name in seen:
+                continue
+            seen.add(name)
+            phase_names.append(name)
+    return phase_names
+
+
+def _count_values(values: list[object]) -> dict:
+    counts: dict[str, int] = {}
+    for value in values:
+        key = str(value or "unknown")
+        counts[key] = int(counts.get(key) or 0) + 1
+    return counts

src/services/readiness/launch_readiness_service.py

@@ -40,9 +40,19 @@ class LaunchReadinessService:
         self.trust_service = trust_service or ModelTrustService()
         self.live_readiness_service = live_readiness_service or LiveReadinessService()
 
-    def build_summary(self, *, instrument_id: int, timeframe: str = "1m") -> dict:
+    def build_summary(
+        self,
+        *,
+        instrument_id: int,
+        timeframe: str = "1m",
+        cached_news_only: bool = False,
+    ) -> dict:
         now = datetime.now(timezone.utc)
-        market_state = self.market_state_service.build_market_state(instrument_id=instrument_id, timeframe=timeframe)
+        market_state = self.market_state_service.build_market_state(
+            instrument_id=instrument_id,
+            timeframe=timeframe,
+            cached_news_only=cached_news_only,
+        )
         broker_state = self.broker_read_model_service.fetch_snapshot(
             instrument_id=instrument_id,
             now=now,

src/web/styles.css

@@ -246,6 +246,20 @@ th { color: #9db0d3; }
   border-color: #4f7cff;
   box-shadow: inset 0 0 0 1px rgba(79, 124, 255, 0.25);
 }
+.assistant-archive-panel {
+  display: grid;
+  gap: 10px;
+  margin-bottom: 10px;
+}
+.assistant-archive-controls {
+  display: grid;
+  grid-template-columns: repeat(auto-fit, minmax(160px, 1fr));
+  gap: 10px;
+}
+.assistant-compare-selected {
+  border-color: #fbbf24;
+  box-shadow: inset 0 0 0 1px rgba(251, 191, 36, 0.28);
+}
 .assistant-dense .card {
   padding: 11px;
 }
@@ -272,6 +286,33 @@ th { color: #9db0d3; }
   border-color: #4f7cff;
   box-shadow: inset 0 0 0 1px rgba(79, 124, 255, 0.3);
 }
+.assistant-preference-panel {
+  display: grid;
+  gap: 10px;
+  margin-bottom: 10px;
+}
+.assistant-preference-fields,
+.assistant-preference-template-grid {
+  display: grid;
+  grid-template-columns: repeat(auto-fit, minmax(160px, 1fr));
+  gap: 10px;
+}
+.assistant-preference-field {
+  display: grid;
+  gap: 6px;
+}
+.assistant-preference-input {
+  width: 100%;
+  padding: 8px 10px;
+  border: 1px solid #31436c;
+  border-radius: 8px;
+  background: #0d1430;
+  color: #fff;
+}
+.assistant-preference-dirty {
+  color: #fcd34d;
+  font-size: 12px;
+}
 .actions {
   display: flex;
   gap: 10px;

tests/unit/test_api_server.py

@@ -173,46 +173,75 @@ class ApiServerRoutingTests(unittest.TestCase):
         )
         digest_archive_status, digest_archive_payload = route_request(
             "/trading/alerts/digest-history-archive",
-            {"instrument_id": ["1"], "limit": ["5"], "offset": ["0"], "preferred_history_kind": ["reminder"]},
+            {"instrument_id": ["1"], "limit": ["5"], "offset": ["0"], "preferred_history_kind": ["reminder"], "query_text": ["candidate"]},
             dependencies={
                 "trading_alert_operator_digest_history_archive": lambda **kwargs: {
                     "status": "active",
                     "instrument_id": kwargs["instrument_id"],
                     "limit": kwargs["limit"],
-                    "filters": {"preferred_history_kind": kwargs["preferred_history_kind"]},
+                    "filters": {"preferred_history_kind": kwargs["preferred_history_kind"], "query_text": kwargs["query_text"]},
+                }
+            },
+        )
+        digest_export_status, digest_export_payload = route_request(
+            "/trading/alerts/digest-history-archive/export",
+            {"instrument_id": ["1"], "query_text": ["candidate"], "left_id": ["4"], "right_id": ["3"]},
+            dependencies={
+                "trading_alert_operator_digest_history_export": lambda **kwargs: {
+                    "status": "active",
+                    "export_kind": "digest_archive_bundle",
+                    "query_text": kwargs["query_text"],
+                    "left_id": kwargs["left_id"],
+                    "right_id": kwargs["right_id"],
                 }
             },
         )
         digest_compare_status, digest_compare_payload = route_request(
             "/trading/alerts/digest-history-archive/compare",
-            {"instrument_id": ["1"], "left_id": ["4"], "right_id": ["3"]},
+            {"instrument_id": ["1"], "left_id": ["4"], "right_id": ["3"], "query_text": ["candidate"]},
             dependencies={
                 "trading_alert_operator_digest_history_compare": lambda **kwargs: {
                     "status": "active",
                     "left_id": kwargs["left_id"],
                     "right_id": kwargs["right_id"],
+                    "query_text": kwargs["query_text"],
                 }
             },
         )
         workflow_archive_status, workflow_archive_payload = route_request(
             "/trading/workflow-archive",
-            {"instrument_id": ["1"], "current_session": ["LONDON"], "next_phase": ["intraday"]},
+            {"instrument_id": ["1"], "current_session": ["LONDON"], "next_phase": ["intraday"], "query_text": ["handoff"]},
             dependencies={
                 "trading_workflow_archive": lambda **kwargs: {
                     "status": "active",
                     "instrument_id": kwargs["instrument_id"],
-                    "filters": {"current_session": kwargs["current_session"], "next_phase": kwargs["next_phase"]},
+                    "filters": {"current_session": kwargs["current_session"], "next_phase": kwargs["next_phase"], "query_text": kwargs["query_text"]},
+                }
+            },
+        )
+        workflow_export_status, workflow_export_payload = route_request(
+            "/trading/workflow-archive/export",
+            {"instrument_id": ["1"], "query_text": ["handoff"], "left_id": ["9"], "right_id": ["8"]},
+            dependencies={
+                "trading_workflow_archive_export": lambda **kwargs: {
+                    "status": "active",
+                    "export_kind": "workflow_archive_bundle",
+                    "query_text": kwargs["query_text"],
+                    "left_id": kwargs["left_id"],
+                    "right_id": kwargs["right_id"],
+                    "handoff_bundle_summary": {"status": "active"},
                 }
             },
         )
         workflow_compare_status, workflow_compare_payload = route_request(
             "/trading/workflow-archive/compare",
-            {"instrument_id": ["1"], "left_id": ["9"], "right_id": ["8"]},
+            {"instrument_id": ["1"], "left_id": ["9"], "right_id": ["8"], "query_text": ["handoff"]},
             dependencies={
                 "trading_workflow_archive_compare": lambda **kwargs: {
                     "status": "active",
                     "left_id": kwargs["left_id"],
                     "right_id": kwargs["right_id"],
+                    "query_text": kwargs["query_text"],
                 }
             },
         )
@@ -248,17 +277,87 @@ class ApiServerRoutingTests(unittest.TestCase):
         self.assertEqual(digest_archive_status, 200)
         self.assertEqual(digest_archive_payload["limit"], 5)
         self.assertEqual(digest_archive_payload["filters"]["preferred_history_kind"], "reminder")
+        self.assertEqual(digest_archive_payload["filters"]["query_text"], "candidate")
+        self.assertEqual(digest_export_status, 200)
+        self.assertEqual(digest_export_payload["export_kind"], "digest_archive_bundle")
+        self.assertEqual(digest_export_payload["left_id"], 4)
         self.assertEqual(digest_compare_status, 200)
         self.assertEqual(digest_compare_payload["left_id"], 4)
+        self.assertEqual(digest_compare_payload["query_text"], "candidate")
         self.assertEqual(workflow_archive_status, 200)
         self.assertEqual(workflow_archive_payload["filters"]["next_phase"], "intraday")
+        self.assertEqual(workflow_archive_payload["filters"]["query_text"], "handoff")
+        self.assertEqual(workflow_export_status, 200)
+        self.assertEqual(workflow_export_payload["export_kind"], "workflow_archive_bundle")
+        self.assertEqual(workflow_export_payload["handoff_bundle_summary"]["status"], "active")
         self.assertEqual(workflow_compare_status, 200)
         self.assertEqual(workflow_compare_payload["right_id"], 8)
+        self.assertEqual(workflow_compare_payload["query_text"], "handoff")
         self.assertEqual(focus_get_status, 200)
         self.assertEqual(focus_get_payload["focus_candidate_id"], 18)
         self.assertEqual(focus_post_status, 200)
         self.assertEqual(focus_post_payload["profile_key"], "desk_a")
 
+    def test_routes_decision_assistant_profile_management(self) -> None:
+        create_status, create_payload = route_request(
+            "/trading/alerts/digest-preferences/profiles",
+            {},
+            method="POST",
+            body={
+                "instrument_id": 1,
+                "profile_key": "desk_b",
+                "profile_label": "Desk B",
+                "digest_mode": "high_value_only",
+            },
+            dependencies={
+                "trading_alert_operator_digest_profiles_create": lambda **kwargs: {
+                    "status": "created",
+                    "instrument_id": kwargs["instrument_id"],
+                    "profile_key": kwargs["profile_key"],
+                    "profile_label": kwargs["profile_label"],
+                }
+            },
+        )
+        rename_status, rename_payload = route_request(
+            "/trading/alerts/digest-preferences/profiles/desk_b/rename",
+            {},
+            method="POST",
+            body={
+                "instrument_id": 1,
+                "profile_label": "Desk Bravo",
+            },
+            dependencies={
+                "trading_alert_operator_digest_profiles_rename": lambda **kwargs: {
+                    "status": "renamed",
+                    "instrument_id": kwargs["instrument_id"],
+                    "profile_key": kwargs["profile_key"],
+                    "profile_label": kwargs["profile_label"],
+                }
+            },
+        )
+        delete_status, delete_payload = route_request(
+            "/trading/alerts/digest-preferences/profiles/desk_b/delete",
+            {},
+            method="POST",
+            body={"instrument_id": 1},
+            dependencies={
+                "trading_alert_operator_digest_profiles_delete": lambda **kwargs: {
+                    "status": "deleted",
+                    "instrument_id": kwargs["instrument_id"],
+                    "profile_key": kwargs["profile_key"],
+                }
+            },
+        )
+
+        self.assertEqual(create_status, 200)
+        self.assertEqual(create_payload["status"], "created")
+        self.assertEqual(create_payload["profile_key"], "desk_b")
+        self.assertEqual(rename_status, 200)
+        self.assertEqual(rename_payload["profile_key"], "desk_b")
+        self.assertEqual(rename_payload["profile_label"], "Desk Bravo")
+        self.assertEqual(delete_status, 200)
+        self.assertEqual(delete_payload["status"], "deleted")
+
     def test_routes_maintenance_summary(self) -> None:
         status, payload = route_request(
             "/trading/maintenance-summary",

tests/unit/test_api_static_server.py

@@ -104,8 +104,11 @@ class ApiStaticServerTests(unittest.TestCase):
             "/trading/alerts/quality-summary",
             "/trading/alerts/operator-digest",
             "/trading/alerts/digest-preferences",
+            "/trading/alerts/digest-preferences/profiles",
             "/trading/alerts/digest-history-archive",
+            "/trading/alerts/digest-history-archive/export",
             "/trading/workflow-archive",
+            "/trading/workflow-archive/export",
             "/trading/assistant-focus",
             "/trading/runtime-incidents",
             "/trading/maintenance-summary",
@@ -130,6 +133,40 @@ class ApiStaticServerTests(unittest.TestCase):
         ):
             self.assertNotIn(forbidden, app_js)
 
+    def test_dashboard_exposes_workflow_archive_operations_surface(self) -> None:
+        app_file = _resolve_static_file("/static/app.js")
+        self.assertIsNotNone(app_file)
+        app_js = app_file.read_text(encoding="utf-8")
+
+        for expected in (
+            "workflow operations",
+            "workflow drift",
+            "Day Group:",
+            "Session Group:",
+            "operation snapshots",
+            "drift next prompt",
+            "Search Text",
+            "Export Archive Results",
+            "Export Archive Compare",
+            "Export Workflow Results",
+            "Export Workflow Compare",
+            "Replay Bridge",
+            "Note Bridge",
+            "Handoff Bridge",
+            "Clear Workflow Bridge",
+            "provider health cards",
+            "Open Visual Bridge",
+            "Open Target",
+            "Open Embed",
+            "archive trend",
+            "archive baseline",
+            "Pin Baseline",
+            "Compare To Baseline",
+            "workflow long-range",
+            "workflow baseline",
+        ):
+            self.assertIn(expected, app_js)
+
 
 if __name__ == "__main__":
     unittest.main()

tests/unit/test_archive_long_range_projection.py

@@ -0,0 +1,92 @@
+import unittest
+
+from src.services.operator.archive_long_range_projection import (
+    build_digest_archive_baseline_summary,
+    build_digest_archive_long_range_summary,
+    build_workflow_archive_baseline_summary,
+    build_workflow_archive_long_range_summary,
+)
+
+
+class ArchiveLongRangeProjectionTests(unittest.TestCase):
+    def test_digest_long_range_and_baseline_summary(self) -> None:
+        items = [
+            {
+                "snapshot_id": 5,
+                "created_ts": "2026-05-14T09:00:00+00:00",
+                "top_summary": "Latest digest",
+                "high_value_reminder_count": 2,
+            },
+            {
+                "snapshot_id": 4,
+                "created_ts": "2026-05-10T09:00:00+00:00",
+                "top_summary": "Weekly digest",
+                "high_value_reminder_count": 1,
+            },
+            {
+                "snapshot_id": 1,
+                "created_ts": "2026-04-20T09:00:00+00:00",
+                "top_summary": "Older digest",
+                "high_value_reminder_count": 0,
+            },
+        ]
+
+        trend = build_digest_archive_long_range_summary(items)
+        baseline = build_digest_archive_baseline_summary(items)
+
+        self.assertEqual(trend["status"], "active")
+        self.assertEqual(trend["recent_7d_count"], 2)
+        self.assertEqual(trend["recent_30d_count"], 3)
+        self.assertGreaterEqual(trend["covered_day_span"], 20)
+        self.assertTrue(trend["next_review_prompt"])
+        self.assertEqual(baseline["status"], "active")
+        self.assertEqual(baseline["latest_snapshot_id"], 5)
+        self.assertEqual(baseline["baseline_snapshot_id"], 1)
+        self.assertTrue(baseline["compare_ready"])
+
+    def test_workflow_long_range_and_baseline_summary(self) -> None:
+        items = [
+            {
+                "snapshot_id": 9,
+                "created_ts": "2026-05-14T09:00:00+00:00",
+                "current_session": "LONDON",
+                "next_phase": "intraday",
+                "ready_for_handoff_count": 2,
+            },
+            {
+                "snapshot_id": 8,
+                "created_ts": "2026-05-08T09:00:00+00:00",
+                "current_session": "NY_AM",
+                "next_phase": "post_trade",
+                "ready_for_handoff_count": 1,
+            },
+            {
+                "snapshot_id": 3,
+                "created_ts": "2026-04-18T09:00:00+00:00",
+                "current_session": "ASIA_PM",
+                "next_phase": "pre_market",
+                "ready_for_handoff_count": 0,
+            },
+        ]
+
+        trend = build_workflow_archive_long_range_summary(items)
+        baseline = build_workflow_archive_baseline_summary(items)
+
+        self.assertEqual(trend["status"], "active")
+        self.assertEqual(trend["recent_7d_count"], 2)
+        self.assertEqual(trend["recent_30d_count"], 3)
+        self.assertEqual(trend["distinct_session_count"], 3)
+        self.assertEqual(baseline["status"], "active")
+        self.assertEqual(baseline["latest_snapshot_id"], 9)
+        self.assertEqual(baseline["baseline_snapshot_id"], 3)
+        self.assertTrue(baseline["suggested_compare_prompt"])
+
+    def test_empty_items_return_idle(self) -> None:
+        self.assertEqual(build_digest_archive_long_range_summary([])["status"], "idle")
+        self.assertEqual(build_digest_archive_baseline_summary([])["status"], "idle")
+        self.assertEqual(build_workflow_archive_long_range_summary([])["status"], "idle")
+        self.assertEqual(build_workflow_archive_baseline_summary([])["status"], "idle")
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/unit/test_documentation_consistency.py

@@ -8,8 +8,6 @@ from scripts import check_local_demo_readiness, check_trading_delivery_readiness
 ROOT = Path(__file__).resolve().parents[2]
 README = ROOT / "README.md"
 RUNBOOK = ROOT / "RUNBOOK.md"
-TRADING_PLAN_DOC = ROOT / "docs" / "AI_ICT_实盘交易分析系统开发计划_v1.md"
-ICT_BOOK_DOC = ROOT / "docs" / "ICT交易全书_出版终校版.md"
 TRAINING_PUBLIC_DOC = ROOT / "docs" / "training_console_public_surface_v1.md"
 TRAINING_PACKAGE_SMOKE_DOC = ROOT / "docs" / "examples" / "training_case_package_smoke_m6.md"
 TRADING_INDEX_HTML = ROOT / "src" / "web" / "index.html"
@@ -18,15 +16,20 @@ TRAINING_HTML = ROOT / "src" / "web" / "training.html"
 TRAINING_JS = ROOT / "src" / "web" / "training.js"
 OLD_TRAINING_DOC = ROOT / "docs" / "training_case_engine_m1_m6_product_readme.md"
 NONEXISTENT_VISUAL_PLAN_NAME = "AI_ICT_" + "视觉盯盘与主动告警增强方案_v1.md"
-ARCHIVED_DORMANT_DOCS = {
-    "AI_ICT_下一阶段开发计划_v1.md",
-    "AI_ICT_下一阶段开发任务清单_v1.md",
-    "AI_ICT_TradingAgents融合实施方案_v1.md",
-    "AI_ICT_方案一系统边界图与模块融合图_v1.md",
+
+ACTIVE_TOP_LEVEL_DOCS = {
+    "AI_ICT_分析助手主线增强开发计划_2026-05-14_v1.md",
+    "AI_ICT_全自动开发执行流程_v1.md",
+    "AI_ICT_实盘交易分析决策助手下一阶段开发方案_v1.md",
+    "AI_ICT_实盘交易分析系统开发计划_v1.md",
+    "AI_ICT_实盘辅助决策系统开发方案_v2.md",
+    "AI_ICT_自动化执行层_P15-PXX闭环任务清单_v1.md",
+    "AI_ICT_需求文档剩余功能与延期项清单_2026-05-14_v1.md",
+    "ICT交易全书_出版终校版.md",
+    "training_console_public_surface_v1.md",
 }
-OPTIONAL_DORMANT_DOCS = {
-    "AI_ICT_repair_bridge_protocol_v1.md",
-    "AI_ICT_research_council脚手架与数据表草案_v1.md",
+
+ARCHIVED_DOCS = {
     "AI_ICT_TradingAgents_P0-P6合理性审查与优化结论_v1.md",
     "AI_ICT_TradingAgents_P0自动任务清单_v1.md",
     "AI_ICT_TradingAgents_P1自动任务清单_v1.md",
@@ -35,44 +38,38 @@ OPTIONAL_DORMANT_DOCS = {
     "AI_ICT_TradingAgents_P4自动任务清单_v1.md",
     "AI_ICT_TradingAgents_P5自动任务清单_v1.md",
     "AI_ICT_TradingAgents_P6自动任务清单_v1.md",
+    "AI_ICT_TradingAgents融合实施方案_v1.md",
     "AI_ICT_TradingAgents融合实施方案_最终版.md",
+    "AI_ICT_下一阶段开发任务清单_v1.md",
+    "AI_ICT_下一阶段开发计划_v1.md",
+    "AI_ICT_repair_bridge_protocol_v1.md",
+    "AI_ICT_research_council脚手架与数据表草案_v1.md",
     "AI_ICT_外部repair_agent接入说明_v1.md",
+    "AI_ICT_实盘交易分析决策助手未完成需求清单_2026-05-14_v1.md",
+    "AI_ICT_方案一系统边界图与模块融合图_v1.md",
     "AI_ICT_盯盘告警开发任务清单_v1.md",
     "AI_ICT_盯盘告警技术设计稿_v1.md",
     "AI_ICT_盯盘告警架构图_v1.md",
-    "AI_ICT_自动化执行层_P15-PXX闭环任务清单_v1.md",
     "AI_ICT_自动化执行层_P7监控成熟度与晋级路由_v1.md",
     "AI_ICT_自动化流程使用教程_v1.md",
 }
 
 
 class DocumentationConsistencyTests(unittest.TestCase):
-    def test_docs_inventory_is_current_and_old_training_product_doc_is_gone(self) -> None:
+    def test_docs_inventory_is_current_and_archived_docs_are_not_top_level(self) -> None:
         top_level_docs = {
             path.name
             for path in (ROOT / "docs").iterdir()
             if path.is_file()
         }
-        self.assertEqual(
-            top_level_docs,
-            {
-                "AI_ICT_全自动开发执行流程_v1.md",
-                "AI_ICT_实盘交易分析决策助手下一阶段开发方案_v1.md",
-                "AI_ICT_实盘交易分析决策助手未完成需求清单_2026-05-14_v1.md",
-                "AI_ICT_实盘交易分析系统开发计划_v1.md",
-                "AI_ICT_实盘辅助决策系统开发方案_v2.md",
-                "ICT交易全书_出版终校版.md",
-                "training_console_public_surface_v1.md",
-            }
-            | OPTIONAL_DORMANT_DOCS,
-        )
-        self.assertTrue(TRAINING_PACKAGE_SMOKE_DOC.is_file())
+        self.assertEqual(top_level_docs, ACTIVE_TOP_LEVEL_DOCS)
+
         archive_docs = {
             path.name
             for path in (ROOT / "docs" / "archive").iterdir()
             if path.is_file()
         }
-        self.assertTrue(ARCHIVED_DORMANT_DOCS <= archive_docs)
+        self.assertTrue(ARCHIVED_DOCS <= archive_docs)
         self.assertFalse(OLD_TRAINING_DOC.exists())
 
     def test_readme_and_runbook_use_current_training_entry_and_not_old_doc(self) -> None:

tests/unit/test_operator_workspace_service.py

@@ -191,7 +191,23 @@ class OperatorWorkspaceServiceTests(unittest.TestCase):
         self.assertIn("workflow_history_archive", summary["workflow_artifact_summary"])
         self.assertIn("workflow_history_archive_count", summary["workflow_artifact_summary"])
         self.assertIn("workflow_archive_compare", summary["workflow_artifact_summary"])
+        self.assertIn("workflow_archive_operations_summary", summary["workflow_artifact_summary"])
+        self.assertIn("workflow_archive_multi_day_drift_review", summary["workflow_artifact_summary"])
+        self.assertIn("workflow_archive_bridge_summary", summary["workflow_artifact_summary"])
         self.assertIn("workflow_persistence_status", summary["workflow_artifact_summary"])
+        self.assertEqual(summary["workflow_artifact_summary"]["workflow_archive_operations_summary"]["status"], "active")
+        self.assertEqual(
+            summary["workflow_artifact_summary"]["workflow_archive_multi_day_drift_review"]["status"],
+            "insufficient_history",
+        )
+        self.assertEqual(summary["workflow_artifact_summary"]["workflow_archive_bridge_summary"]["status"], "active")
+        self.assertIn("long_range_trend_summary", summary["workflow_artifact_summary"]["workflow_history_archive"])
+        self.assertIn("baseline_pin_summary", summary["workflow_artifact_summary"]["workflow_history_archive"])
+        self.assertEqual(summary["workflow_artifact_summary"]["workflow_history_archive"]["long_range_trend_summary"]["status"], "active")
+        self.assertEqual(summary["workflow_artifact_summary"]["workflow_history_archive"]["baseline_pin_summary"]["status"], "insufficient_history")
+        self.assertIn("daily_groups", summary["workflow_artifact_summary"]["workflow_history_archive"])
+        self.assertIn("session_groups", summary["workflow_artifact_summary"]["workflow_history_archive"])
+        self.assertIn("bridge_summary", summary["workflow_artifact_summary"]["workflow_history_archive"])
         self.assertIn("assistant_loop_queue", summary)
         self.assertIn("assistant_loop_summary", summary)
         self.assertTrue(summary["assistant_loop_queue"])
@@ -223,6 +239,10 @@ class OperatorWorkspaceServiceTests(unittest.TestCase):
         self.assertEqual(summary["decision_timeline"][0]["title"], "Current Packet")
         self.assertEqual(summary["evidence_chain"][0]["title"], "Decision Gate")
         self.assertEqual(summary["evidence_chain"][1]["title"], "Launch Mode")
+        self.assertTrue(any(item["title"] == "Focus Candidate" for item in summary["evidence_chain"]))
+        focus_evidence = next(item for item in summary["evidence_chain"] if item["title"] == "Focus Candidate")
+        self.assertTrue(focus_evidence["focus_related"])
+        self.assertEqual(focus_evidence["focus_candidate_id"], 18)
         self.assertEqual(summary["explanation_drilldown"][0]["title"], "Decision Logic")
         self.assertEqual(summary["reason_citations"][0]["source_path"], "decision_packet.why_allowed[0]")
         self.assertEqual(summary["learning_rollup"]["status"], "active")
@@ -230,6 +250,9 @@ class OperatorWorkspaceServiceTests(unittest.TestCase):
         self.assertEqual(summary["review_entrypoints"]["status"], "active")
         self.assertEqual(summary["review_entrypoints"]["focus_candidate_id"], 18)
         self.assertGreaterEqual(summary["review_entrypoints"]["focus_note_count"], 1)
+        self.assertIn("Candidate #18", summary["review_entrypoints"]["focus_note_prefill"])
+        self.assertIn("focus-candidate", summary["review_entrypoints"]["focus_note_tags"])
+        self.assertIn("focus_note_prompt", summary["review_entrypoints"])
         self.assertEqual(summary["review_entrypoints"]["actions"][0]["action_id"], "open_operator_notes")
         self.assertTrue(summary["review_entrypoints"]["note_templates"])
         self.assertIn("review_entry_action_queue", summary["review_entrypoints"])
@@ -368,8 +391,15 @@ class OperatorWorkspaceServiceTests(unittest.TestCase):
         self.assertIn("digest_history_archive", summary["digest_surface"])
         self.assertIn("digest_history_archive_count", summary["digest_surface"])
         self.assertIn("digest_archive_compare", summary["digest_surface"])
+        self.assertIn("focus_related_reminders", summary["digest_surface"])
+        self.assertGreaterEqual(summary["digest_surface"]["focus_related_reminder_count"], 1)
+        self.assertIsNotNone(summary["digest_surface"]["next_focus_related_reminder"])
         self.assertIn("digest_persistence_status", summary["digest_surface"])
         self.assertIn("next_digest_history_archive_review", summary["digest_surface"])
+        self.assertIn("long_range_trend_summary", summary["digest_surface"]["digest_history_archive"])
+        self.assertIn("baseline_pin_summary", summary["digest_surface"]["digest_history_archive"])
+        self.assertEqual(summary["digest_surface"]["digest_history_archive"]["long_range_trend_summary"]["status"], "active")
+        self.assertEqual(summary["digest_surface"]["digest_history_archive"]["baseline_pin_summary"]["status"], "insufficient_history")
         self.assertIn("next_high_value_change", summary["digest_surface"])
         self.assertIn("digest_review_back_actions", summary["digest_surface"])
         self.assertIn("digest_review_back_action_count", summary["digest_surface"])
@@ -427,6 +457,7 @@ class OperatorWorkspaceServiceTests(unittest.TestCase):
         self.assertTrue(summary["digest_history_preview"])
         self.assertEqual(summary["digest_history_preview"][0]["kind"], "reminder")
         self.assertEqual(summary["personalized_reminders"][0]["severity"], "high")
+        self.assertTrue(any(item.get("focus_related") for item in summary["personalized_reminders"]))
         self.assertIn("importance_score", summary["personalized_reminders"][0])
         self.assertIn("importance_template", summary["personalized_reminders"][0])
         self.assertIn("high_value", summary["personalized_reminders"][0])
@@ -434,6 +465,9 @@ class OperatorWorkspaceServiceTests(unittest.TestCase):
         self.assertEqual(summary["setup_learning_rollup"]["top_lessons"][0]["lesson_type"], "journal_pattern")
         self.assertEqual(summary["setup_learning_rollup"]["session_code"], "LONDON")
         self.assertEqual(summary["setup_learning_rollup"]["session_code_source"], "focus_candidate")
+        self.assertEqual(summary["setup_learning_rollup"]["focus_candidate_id"], 18)
+        self.assertEqual(summary["setup_learning_rollup"]["focus_replay_scope"], "candidate_scoped")
+        self.assertIn("focus_replay_prompt", summary["setup_learning_rollup"])
         self.assertIn("replay_statistics", summary["setup_learning_rollup"])
         self.assertIn("learning_attribution_summary", summary["setup_learning_rollup"])
         self.assertIn("failure_reason_attribution", summary["setup_learning_rollup"])
@@ -591,6 +625,8 @@ class OperatorWorkspaceServiceTests(unittest.TestCase):
         self.assertTrue(summary["external_context_summary"]["announcement_titles"])
         self.assertIn("macro_news_calendar", summary["external_context_summary"])
         self.assertGreaterEqual(summary["external_context_summary"]["macro_news_calendar_count"], 3)
+        self.assertIn("macro_news_provider_inventory", summary["external_context_summary"])
+        self.assertGreaterEqual(summary["external_context_summary"]["macro_news_provider_count"], 1)
         self.assertIn("event_window_summary", summary["external_context_summary"])
         self.assertIn("event_window_status", summary["external_context_summary"])
         self.assertIn("event_window_prompt", summary["external_context_summary"])
@@ -598,11 +634,24 @@ class OperatorWorkspaceServiceTests(unittest.TestCase):
         self.assertTrue(summary["external_context_summary"]["exchange_anomaly_window"])
         self.assertIn("window_type", summary["external_context_summary"]["macro_news_calendar"][0])
         self.assertIn("chart_readonly_sources", summary["external_context_summary"])
+        self.assertIn("chart_provider_registry", summary["external_context_summary"])
+        self.assertGreaterEqual(summary["external_context_summary"]["chart_provider_registry_count"], 3)
         self.assertIn("chart_stale_source_evidence", summary["external_context_summary"])
         self.assertGreaterEqual(summary["external_context_summary"]["chart_readonly_source_count"], 1)
         self.assertEqual(summary["external_context_summary"]["chart_stale_source_count"], 0)
         self.assertEqual(summary["external_context_summary"]["chart_adapter_mode"], "live_reference")
         self.assertEqual(summary["external_context_summary"]["chart_primary_mode"], "internal_review_packet_html")
+        self.assertIn("exchange_context_sources", summary["external_context_summary"])
+        self.assertGreaterEqual(summary["external_context_summary"]["exchange_context_source_count"], 3)
+        self.assertIn("provider_health_cards", summary["external_context_summary"])
+        self.assertGreaterEqual(summary["external_context_summary"]["provider_health_card_count"], 3)
+        self.assertIn("provider_health_summary", summary["external_context_summary"])
+        self.assertEqual(summary["external_context_summary"]["provider_health_summary"]["status"], "active")
+        self.assertIn("visual_evidence_bridge_summary", summary["external_context_summary"])
+        self.assertEqual(summary["external_context_summary"]["visual_evidence_bridge_summary"]["status"], "active")
+        self.assertIn("visual_reference_cards", summary["external_context_summary"])
+        self.assertGreaterEqual(summary["external_context_summary"]["visual_reference_card_count"], 3)
+        self.assertIsNotNone(summary["external_context_summary"]["visual_reference_next_prompt"])
         self.assertIn("live", summary["external_context_summary"]["chart_context_prompt"].lower())
         self.assertIn("CPI", summary["external_context_summary"]["headline"])
         self.assertTrue(summary["no_trade_reason_items"])

tests/unit/test_run_until_blocked.py

@@ -140,7 +140,7 @@ class RunUntilBlockedTests(unittest.TestCase):
             changed_files=[
                 "scripts/run_until_blocked.py",
                 "tests/unit/test_run_until_blocked.py",
-                "docs/AI_ICT_自动化流程使用教程_v1.md",
+                "docs/archive/AI_ICT_自动化流程使用教程_v1.md",
                 "SESSION.md",
                 "TASKS.md",
             ],
@@ -1455,3 +1455,4 @@ class RunUntilBlockedTests(unittest.TestCase):
 
 if __name__ == "__main__":
     unittest.main()
+

tests/unit/test_workflow_archive_projection.py

@@ -0,0 +1,88 @@
+import unittest
+
+from src.services.operator.workflow_archive_projection import build_workflow_archive_operations_view
+
+
+class WorkflowArchiveProjectionTests(unittest.TestCase):
+    def test_builds_daily_session_and_drift_views(self) -> None:
+        items = [
+            {
+                "snapshot_id": 3,
+                "created_ts": "2026-05-14T09:30:00+00:00",
+                "current_session": "LONDON",
+                "next_phase": "intraday",
+                "next_artifact_kind": "intraday_digest",
+                "artifact_count": 4,
+                "ready_for_handoff_count": 2,
+                "needs_update_count": 1,
+                "pending_capture_count": 1,
+                "workflow_body": "London plan ready.",
+                "next_artifact_prompt": "Check intraday follow-through.",
+                "phase_briefs": [
+                    {"phase": "pre_market", "archive_body": "Macro and session prep done."},
+                    {"phase": "intraday", "archive_body": "Watch the first liquidity sweep."},
+                ],
+            },
+            {
+                "snapshot_id": 2,
+                "created_ts": "2026-05-13T15:45:00+00:00",
+                "current_session": "NY_AM",
+                "next_phase": "post_trade",
+                "next_artifact_kind": "post_trade_recap",
+                "artifact_count": 2,
+                "ready_for_handoff_count": 1,
+                "needs_update_count": 1,
+                "pending_capture_count": 0,
+                "workflow_body": "NY recap pending.",
+                "next_artifact_prompt": "Write post-trade recap.",
+                "phase_briefs": [
+                    {"phase": "pre_market", "archive_body": "Prep was lighter."},
+                    {"phase": "intraday", "archive_body": "NY sweep already happened."},
+                ],
+            },
+            {
+                "snapshot_id": 1,
+                "created_ts": "2026-05-13T12:15:00+00:00",
+                "current_session": "LONDON",
+                "next_phase": "intraday",
+                "next_artifact_kind": "intraday_digest",
+                "artifact_count": 3,
+                "ready_for_handoff_count": 1,
+                "needs_update_count": 1,
+                "pending_capture_count": 1,
+                "workflow_body": "Earlier London handoff.",
+                "next_artifact_prompt": "Re-check the London setup.",
+                "phase_briefs": [
+                    {"phase": "pre_market", "archive_body": "Prep baseline."},
+                    {"phase": "intraday", "archive_body": "Initial sweep watch."},
+                ],
+            },
+        ]
+
+        result = build_workflow_archive_operations_view(items)
+
+        self.assertEqual(result["daily_group_count"], 2)
+        self.assertEqual(result["session_group_count"], 2)
+        self.assertEqual(result["operations_summary"]["status"], "active")
+        self.assertEqual(result["operations_summary"]["latest_group_date"], "2026-05-14")
+        self.assertEqual(result["operations_summary"]["most_active_session"], "LONDON")
+        self.assertEqual(result["multi_day_drift_review"]["status"], "active")
+        self.assertEqual(result["multi_day_drift_review"]["drift_status"], "drift_detected")
+        self.assertEqual(result["bridge_summary"]["status"], "active")
+        self.assertEqual(result["bridge_summary"]["latest_snapshot_id"], 3)
+        self.assertGreaterEqual(result["multi_day_drift_review"]["changed_phase_count"], 1)
+        self.assertTrue(result["daily_groups"][0]["phase_names"])
+        self.assertEqual(result["session_groups"][0]["session_code"], "LONDON")
+
+    def test_handles_empty_archive(self) -> None:
+        result = build_workflow_archive_operations_view([])
+
+        self.assertEqual(result["daily_group_count"], 0)
+        self.assertEqual(result["session_group_count"], 0)
+        self.assertEqual(result["operations_summary"]["status"], "idle")
+        self.assertEqual(result["multi_day_drift_review"]["status"], "insufficient_history")
+        self.assertEqual(result["bridge_summary"]["status"], "idle")
+
+
+if __name__ == "__main__":
+    unittest.main()