Surface city-card decision reasons before the paragraph

City decision cards already had enough evidence, but the first screen still forced users to read the longer explanation before seeing why a card mattered. The header now caps status chips at three high-priority signals and uses product-facing labels for observed breakouts, stale METARs, AI loading, missing market prices, strong model agreement, and next-report waits.

Constraint: The card should stay lightweight and avoid another explanatory section.
Rejected: Keep six mixed freshness chips in the header | too much noise for first-glance scanning.
Confidence: high
Scope-risk: narrow
Reversibility: clean
Tested: npm run build
Not-tested: Browser visual QA across all city states.

CHANGELOG.md

@@ -3,6 +3,7 @@
 ## 1.5.5 - 2026-04-27
 
 - 新增 Qingdao / 青岛城市，结算锚点接入 Wunderground 青岛胶东国际机场 `ZSQD` 历史页，并补齐别名、时区、预热、官方来源和前端地区归类
+- 城市决策卡顶部状态标签收口为 2-3 个高优先级信号，优先展示“实测突破 / 峰值窗口已过 / METAR 过旧 / AI 解读中 / 市场价格缺失 / 模型高度一致 / 需要等待下一报文”，让用户第一眼看到重点
 - 城市决策卡新增 AI 机场报文解读缓存说明：页面内存缓存保留 loading / 流式片段 / 最终结果，`localStorage` 保存最终成功 payload，后端 AI 缓存不再因 `local_time` 变化失效
 - 城市决策卡兜底文案明确标记“快速证据模式”，避免在 DeepSeek 未完整返回时误写成“AI 机场报文解读正常”
 - 城市决策卡流式 AI 解读改为只请求 METAR/官方观测核心解读与判断依据，最高温中枢、模型一致性和风险清单由后端规则补齐，减少等待时间

frontend/components/dashboard/Dashboard.module.css

@@ -11334,7 +11334,7 @@
   display: flex;
   flex-wrap: wrap;
   gap: 7px;
-  margin-top: 10px;
+  margin: -2px 0 10px;
 }
 
 .root :global(.scan-ai-city-status-tag) {

frontend/components/dashboard/scan-terminal/AiPinnedForecastView.tsx

@@ -353,6 +353,15 @@ function AiPinnedCityCard({
         String((row as { window_phase?: string | null } | null)?.window_phase || "").toLowerCase(),
       ),
   );
+  const modelSpread = modelMax != null && modelMin != null ? modelMax - modelMin : null;
+  const modelHighlyConsistent =
+    modelEntries.length >= 4 && modelSpread != null && modelSpread <= 2;
+  const needsNextBulletin =
+    !observationStale &&
+    !observedHighBreak &&
+    !observedLowBreak &&
+    !peakHasPassed &&
+    (observedLowLag || paceTone === "neutral" || aiForecast.status === "loading");
   const observationFreshnessLabel = buildObservationFreshnessLabel({
     detail,
     displayTime: metarReportTimeDisplay,
@@ -410,43 +419,59 @@ function AiPinnedCityCard({
   const statusTags = uniqueStatusTags([
     observedHighBreak
       ? {
-          label: isEn ? "Observed above models" : "实测突破模型",
+          label: isEn ? "Observed breakout" : "实测突破",
           tone: "red",
         }
       : null,
-    observedLowBreak
-      ? {
-          label: isEn ? "Peak revised down" : "峰值下修",
-          tone: "blue",
-        }
-      : null,
-    observedLowLag
-      ? {
-          label: isEn ? "Warm-up needs proof" : "等待升温确认",
-          tone: "amber",
-        }
-      : null,
-    observationStale
-      ? {
-          label: isEn ? "Stale observation" : "观测过旧",
-          tone: "amber",
-        }
-      : null,
     peakHasPassed
       ? {
           label: isEn ? "Peak window passed" : "峰值窗口已过",
           tone: "muted",
         }
       : null,
-    {
+    observationStale
-      label: aiStatusLabel,
+      ? {
-      tone: aiStatusTone,
+          label: isEn
-    },
+            ? isHkoObservation
-    {
+              ? "HKO stale"
-      label: marketFreshnessLabel,
+              : "METAR stale"
-      tone: marketStatusTone,
+            : isHkoObservation
-    },
+              ? "观测过旧"
-  ]).slice(0, 6);
+              : "METAR 过旧",
+          tone: "amber",
+        }
+      : null,
+    observedLowBreak
+      ? {
+          label: isEn ? "Peak revised down" : "峰值下修",
+          tone: "blue",
+        }
+      : null,
+    aiForecast.status === "loading"
+      ? {
+          label: isEn ? "AI reading" : "AI 解读中",
+          tone: aiStatusTone,
+        }
+      : null,
+    marketDecisionView.status === "unavailable"
+      ? {
+          label: isEn ? "Market missing" : "市场价格缺失",
+          tone: marketStatusTone,
+        }
+      : null,
+    modelHighlyConsistent
+      ? {
+          label: isEn ? "Models agree" : "模型高度一致",
+          tone: "green",
+        }
+      : null,
+    observedLowLag || needsNextBulletin
+      ? {
+          label: isEn ? "Wait next report" : "需要等待下一报文",
+          tone: "amber",
+        }
+      : null,
+  ]).slice(0, 3);
   const localizedRisksRaw =
     (isEn ? aiCityForecast?.risks_en : aiCityForecast?.risks_zh) || [];
   const localizedRisks = Array.isArray(localizedRisksRaw)
@@ -475,6 +500,16 @@ function AiPinnedCityCard({
             {isEn ? "Deep analysis" : "城市深度分析"}
           </span>
           <h3>{displayName}</h3>
+          <div className="scan-ai-city-status-tags">
+            {statusTags.map((tag) => (
+              <span
+                key={tag.label}
+                className={clsx("scan-ai-city-status-tag", tag.tone)}
+              >
+                {tag.label}
+              </span>
+            ))}
+          </div>
           <div className="scan-ai-city-pills">
             <span>{detail?.local_time || row?.local_time || "--"}</span>
             <span>
@@ -486,16 +521,6 @@ function AiPinnedCityCard({
             <span>{isEn ? "Model" : "模型"} {modelRange}</span>
             <span>{isEn ? "Peak" : "峰值"} {peakWindow}</span>
           </div>
-          <div className="scan-ai-city-status-tags">
-            {statusTags.map((tag) => (
-              <span
-                key={tag.label}
-                className={clsx("scan-ai-city-status-tag", tag.tone)}
-              >
-                {tag.label}
-              </span>
-            ))}
-          </div>
           <div className="scan-ai-city-freshness">
             <span>{observationFreshnessLabel}</span>
             <span>{marketFreshnessLabel}</span>