feat(agent-workspace): add active atom summary cards

Author: Jacobinwwey

docs/diataxis/en/explanation/agent-conversation-focus-mode-plan.md

@@ -461,10 +461,11 @@ It is conversation-product hardening on top of the shipped shell.
 `M8.23` now exposes operator-facing diagnostics counts for managed conversation actions and missing correction state.
 `M8.24` now ships the active-atom quick-action rail, active-card highlighting, and active-atom routing for Focus / Learning Path actions.
 `M8.25` now adds active-atom continuity notes and path-target diagnostics so focus/path drift is explicit instead of implicit.
+`M8.26` now adds capability-taxonomy summary cards and latest study-action feedback so the current atom reads like a result surface instead of only a button cluster.
 
 Priority order:
 
-1. richer current-atom result cards and study-loop summaries without reintroducing endpoint-specific UI branching,
+1. cross-turn current-atom result history and deeper study-loop continuity,
 2. only then deeper personalization work,
 3. markdown-reader improvements only when they clearly compound the same study loop.
 

docs/diataxis/en/explanation/development-progress-dashboard.md

@@ -18,9 +18,9 @@ Current top priority is not feature-count expansion.
 Current top priority is **conversation-product hardening on top of the already-green backend baseline**:
 
 - keep backend sufficiency as a standing decision gate rather than an active execution lane,
-- keep closing the correction/product/operator gap exposed after `M8.20-M8.25`,
-- treat `M8.23` as the operator-visibility baseline for conversation-memory/correction actions, `M8.24` as the active-atom action-surface baseline, and `M8.25` as the active-atom pane-state continuity baseline,
-- move the remaining near-term budget into richer current-atom result cards and study-loop summaries instead of reopening backend scope,
+- keep closing the correction/product/operator gap exposed after `M8.20-M8.26`,
+- treat `M8.23` as the operator-visibility baseline for conversation-memory/correction actions, `M8.24` as the active-atom action-surface baseline, `M8.25` as the active-atom pane-state continuity baseline, and `M8.26` as the active-atom study-loop summary baseline,
+- move the remaining near-term budget into cross-turn current-atom result history and deeper study-loop continuity instead of reopening backend scope,
 - keep remote `main` CI interpretation evidence-based: the latest fully completed `main` workflow set is green, and new pushes should be treated as pending until their workflow set finishes.
 
 ## Current Mainline Snapshot (2026-04-21 M8.18)
@@ -172,6 +172,23 @@ Execution anchor:
   - drift messaging when focus changes away from the current docked path target,
   - EN/ZH re-render coverage for the new continuity labels and note strings.
 
+## Latest Mainline Increment (2026-04-21 M8.26 Active-Atom Summary Cards)
+
+- Upgraded the active-atom rail from state-only messaging to a richer result-card surface:
+  - a study-loop summary card now groups current-atom capabilities by a frontend taxonomy instead of per-endpoint branching,
+  - a support-surface summary card now exposes memory vs diagnostic capability counts for the current atom,
+  - the study-loop card now reflects the latest successful current-atom study action after execution.
+- Kept the slice intentionally runtime-local:
+  - no backend route, API contract, or persistence model changed,
+  - the implementation reuses capability descriptors and existing runtime diagnostics events already present on `main`.
+- Closed the next product gap after `M8.25`:
+  - the user no longer has to infer whether the current atom is study-ready from raw button lists alone,
+  - the rail now summarizes the study surface before action execution and reflects the latest study action after execution.
+- Expanded regression coverage for:
+  - capability-taxonomy summary cards without endpoint-specific UI branching,
+  - latest study-action summary updates after execution,
+  - EN/ZH re-render coverage for the new summary headings and metrics.
+
 ## Latest Mainline Increment (2026-04-14 M1 Baseline)
 
 - Added first `AgentConversationAPI` contract surface:
@@ -1275,8 +1292,8 @@ Note: foundation-lane gates now exist on `main`; keep them in the normal verific
 
 1. Treat backend sufficiency as closed-by-default while the gate remains green; do not reopen heavier backend work without new measured pressure.
 2. Treat remote `main` CI as evidence-bound to the latest fully completed workflow set; only reopen CI-triage narrative when a fresh failing run exists.
-3. Make the next planning round about post-`M8.25` bounded conversation-product hardening:
-   - richer current-atom result cards and study-loop summaries without reintroducing endpoint-specific UI branching,
+3. Make the next planning round about post-`M8.26` bounded conversation-product hardening:
+   - cross-turn current-atom result history and deeper study-loop continuity before deeper personalization work,
    - only then deeper personalization work,
    - markdown-reader improvements only when they compound the same study loop.
 4. Treat notification-threshold governance as closed after `M7.29`; do not plan `M7.30+` unless a materially new operator requirement appears.

docs/diataxis/en/explanation/local-backend-sufficiency-and-escalation-plan.md

@@ -29,7 +29,7 @@ First post-checkpoint product-memory slice now landed on `main`:
 - the conversation lane now exposes read-only deterministic managed-state inspection with explicit present/missing key visibility for the selected atom,
 - managed mutation remains session-scoped by design.
 - operator diagnostics now summarize memory-policy execution by layer and operation, so the new product-memory lane is inspectable instead of opaque.
-- the current conversation shell now also ships active-atom focus/path action routing plus continuity notes/path-target diagnostics in the frontend without widening backend scope.
+- the current conversation shell now also ships active-atom focus/path action routing, continuity notes/path-target diagnostics, and capability-taxonomy summary cards in the frontend without widening backend scope.
 
 ## Current Implementation State
 
@@ -247,8 +247,8 @@ Recommended next product order:
    - extend explicit conversation memory APIs beyond the current snapshot/read/managed-key inspection and note/correction/eviction surface,
    - the minimum correction / feedback closure is now integrated through explicit next-step guidance on missing managed state,
    - operator-facing visibility for conversation-memory and correction actions is now integrated through persisted diagnostics summary/index counts,
-   - stronger agent action surfaces around focus mode and learning path are now integrated through the active-atom quick-action rail, toolbar routing, and continuity notes/path-target diagnostics,
-   - next prioritize richer current-atom result cards and study-loop summaries before deeper personalization work.
+   - stronger agent action surfaces around focus mode and learning path are now integrated through the active-atom quick-action rail, toolbar routing, continuity notes/path-target diagnostics, and capability-taxonomy summary cards,
+   - next prioritize cross-turn current-atom result history and deeper study-loop continuity before deeper personalization work.
 2. markdown-reader improvement:
    - only when it clearly compounds knowledge review or focus-mode study loops.
 

docs/diataxis/zh/explanation/agent-conversation-focus-mode-plan.md

@@ -449,10 +449,11 @@
 `M8.23` 现在又补上了托管对话动作与缺失 correction state 的 operator-facing diagnostics 计数。
 `M8.24` 现在已补上当前 atom quick-action rail、active-card 高亮，以及 Focus / Learning Path 对当前 atom 的动作路由。
 `M8.25` 现在又补上当前 atom continuity note 与 path-target diagnostics，让 focus/path drift 从隐式状态变成显式提示。
+`M8.26` 现在又补上 capability-taxonomy summary card 与最近 study-action 反馈，让当前 atom 更像结果面，而不只是动作按钮堆叠。
 
 优先顺序：
 
-1. 继续加固更丰富的当前 atom result card 与 study-loop summary，同时不重新引入 endpoint-specific UI branching，
+1. 继续加固跨轮次的当前 atom result history 与更深的 study-loop continuity，
 2. 之后才进入更深的 personalization 工作，
 3. Markdown 阅读器提升只有在能明确复利同一学习闭环时才进入前列。
 

docs/diataxis/zh/explanation/development-progress-dashboard.md

@@ -18,9 +18,9 @@
 当前重心是 **建立在已转绿 backend 基线之上的 conversation-product 加固**：
 
 - 让 backend sufficiency 继续作为常驻 decision gate，而不是继续占用主动执行预算，
-- 持续收口 `M8.20-M8.25` 之后暴露出的 correction / product / operator 缺口，
-- 将 `M8.23` 视作 conversation-memory / correction action 的 operator-visibility 基线，将 `M8.24` 视作当前 atom action-surface 基线，将 `M8.25` 视作当前 atom pane-state continuity 基线，
-- 把剩余近端预算转向更丰富的当前 atom result card 与 study-loop summary，而不是重开 backend 范围，
+- 持续收口 `M8.20-M8.26` 之后暴露出的 correction / product / operator 缺口，
+- 将 `M8.23` 视作 conversation-memory / correction action 的 operator-visibility 基线，将 `M8.24` 视作当前 atom action-surface 基线，将 `M8.25` 视作当前 atom pane-state continuity 基线，将 `M8.26` 视作当前 atom study-loop summary 基线，
+- 把剩余近端预算转向跨轮次的当前 atom result history 与更深的 study-loop continuity，而不是重开 backend 范围，
 - 远端 `main` CI 解释继续基于代码事实：最近一组已完整结束的主线工作流为绿色，新推送在整组工作流结束前只应视为 pending。
 
 ## 当前主线快照（2026-04-21 M8.18）
@@ -172,6 +172,23 @@
   - focus 偏离当前 path target 时的 drift 消息，
   - EN/ZH 对新 continuity label 与 note 文案的重渲覆盖。
 
+## 主线最新增量（2026-04-21 M8.26 当前 Atom Summary Cards）
+
+- 已把当前 atom rail 从状态提示进一步升级为更丰富的 result-card surface：
+  - study-loop summary card 现在会按前端 capability taxonomy 汇总当前 atom 能力，而不是回到按 endpoint 分支的 UI，
+  - support-surface summary card 现在会暴露当前 atom 的 memory / diagnostic 能力计数，
+  - study-loop card 在动作执行后，还会回显当前 atom 最近一次成功的 study action。
+- 范围继续保持在运行时本地：
+  - 没有新增 backend route、API contract 或持久化模型，
+  - 只复用主线现有 capability descriptor 与 runtime diagnostics event。
+- 已收口 `M8.25` 之后的下一个产品缺口：
+  - 用户不再需要只靠原始按钮列表猜测当前 atom 是否具备 study-ready surface，
+  - rail 现在会在执行前给出 study surface 摘要，并在执行后回显最近的 study action。
+- 回归覆盖同步扩展到：
+  - 不引入 endpoint-specific UI branching 的 capability-taxonomy summary cards，
+  - 执行后的最近 study-action summary 更新，
+  - EN/ZH 对新 summary heading 与 metric 文案的重渲覆盖。
+
 ## 主线最新增量（2026-04-14 M1 基线）
 
 - 已新增第一版 `AgentConversationAPI` 合同面：
@@ -1278,8 +1295,8 @@ npm test -- src/learning/KnowledgeLearningPlatform.persistence.test.ts --runInBa
 
 1. backend sufficiency 在 gate 继续为绿时默认视为已收口；没有新的 measured pressure 前，不重开更重 backend 工作。
 2. 远端 `main` CI 继续以最近一组已完整结束的工作流为事实依据；只有出现新的失败 run，才重开 CI-triage 叙事。
-3. 下一轮 planning 应只讨论 post-`M8.25` 的有界 conversation-product 加固：
-   - 更丰富的当前 atom result card 与 study-loop summary，同时不重新引入 endpoint-specific UI branching，
+3. 下一轮 planning 应只讨论 post-`M8.26` 的有界 conversation-product 加固：
+   - 跨轮次的当前 atom result history 与更深的 study-loop continuity，
    - 之后才进入更深的 personalization 工作，
    - Markdown 阅读器提升只有在能复利同一学习闭环时才进入前列。
 4. notification-threshold 治理链路继续视为在 `M7.29` 后已收口；没有新的强运维需求前，不再规划 `M7.30+`。

docs/diataxis/zh/explanation/local-backend-sufficiency-and-escalation-plan.md

@@ -29,7 +29,7 @@
 - 对话 lane 现在也已暴露针对所选 atom 的只读确定性托管状态检查，并能显式区分已存在/缺失的托管 key，
 - 托管写入/修正/清理仍然按设计限制在 session 层。
 - operator diagnostics 现在会按 layer / operation 汇总 memory-policy 执行情况，使这条 product-memory 主线可观测、不可黑箱化。
-- 当前对话 shell 也已经在不扩大 backend 范围的前提下，补上围绕当前 atom 的 focus/path 动作路由，以及 continuity note / path-target diagnostics。
+- 当前对话 shell 也已经在不扩大 backend 范围的前提下，补上围绕当前 atom 的 focus/path 动作路由、continuity note / path-target diagnostics，以及 capability-taxonomy summary card。
 
 ## 当前实现状态
 
@@ -246,8 +246,8 @@
    - 在当前 snapshot/read/托管 key 检查与 note/correction/eviction 能力之上继续扩展显式 conversation memory API，
    - 最小 correction / feedback 收口已经通过“缺失托管状态 -> 显式下一步提示”集成到主线，
    - conversation-memory / correction action 的 operator-facing visibility 已经通过持久化 diagnostics summary / index 计数集成到主线，
-   - 围绕 focus mode 与 learning path 的更强 agent action surface 已经通过当前 atom quick-action rail、toolbar routing 与 continuity note / path-target diagnostics 集成到主线，
-   - 下一步是在更深 personalization 之前，优先补上更丰富的当前 atom result card 与 study-loop summary。
+   - 围绕 focus mode 与 learning path 的更强 agent action surface 已经通过当前 atom quick-action rail、toolbar routing、continuity note / path-target diagnostics，以及 capability-taxonomy summary card 集成到主线，
+   - 下一步是在更深 personalization 之前，优先补上跨轮次的当前 atom result history 与更深的 study-loop continuity。
 2. markdown reader 提升：
    - 只有在它能明确加强知识 review 或 focus-mode 学习闭环时，再提升优先级。