fix(openclaw): DockerContainerUnhealthy + TargetDown 專屬規則匹配

- DockerContainerUnhealthy: ssh docker inspect + docker restart，含 healthcheck 指令驗證
- TargetDown / IP:port instance: ssh 確認 exporter 存活
- 修正 target 混用 alertname 作為 deployment 名稱的問題
- alertname/labels 從 alert_context 提取供規則判斷
- 2026-04-09 ogt: 新增兩條專屬規則

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

apps/api/src/services/openclaw.py

@@ -584,6 +584,8 @@ class OpenClawService:
         raw_namespace = alert_context.get("namespace", "default")
         message = alert_context.get("message", "")
         metrics = alert_context.get("metrics", {})
+        labels = alert_context.get("labels", {})
+        alertname = labels.get("alertname", alert_type)
 
         # Phase 18.1: 正規化資源名稱 (ADR-016)
         # 確保 kubectl 指令使用有效的 K8s 名稱
@@ -622,7 +624,75 @@ class OpenClawService:
 
         # 根據告警類型生成專業 RCA + 仲裁
         # 🔴 2026-03-29 ogt: Mock 響應必須標示為規則匹配，不是 AI 仲裁
-        if "oom" in message.lower() or "memory" in alert_type.lower():
+        # 2026-04-09 ogt: 新增 DockerContainerUnhealthy / TargetDown 專屬規則
+        if alertname == "DockerContainerUnhealthy" or "unhealthy" in message.lower():
+            container_name = labels.get("name", target)
+            host = labels.get("instance", "").split(":")[0] or "192.168.0.188"
+            mock_response = {
+                "action_title": f"檢查 Docker 容器 {container_name} 健康狀態",
+                "description": f"⚙️ 規則匹配: Docker 容器 {container_name} ({host}) healthcheck 失敗。常見原因: 應用程式啟動慢、healthcheck 指令錯誤、依賴服務未就緒。{signoz_summary}",
+                "suggested_action": "RESTART_DEPLOYMENT",
+                "kubectl_command": f"ssh {host} 'docker inspect {container_name} --format=\"{{{{.State.Health.Status}}}}\" && docker restart {container_name}'",
+                "target_resource": container_name,
+                "namespace": namespace,
+                "risk_level": "medium",
+                "blast_radius": {
+                    "affected_pods": 1,
+                    "estimated_downtime": "~30s",
+                    "related_services": [container_name],
+                    "data_impact": "NONE",
+                },
+                "primary_responsibility": "INFRA",
+                "responsibility_reasoning": "Docker 容器健康檢查失敗屬基礎設施團隊責任，需確認 healthcheck 設定與容器狀態",
+                "secondary_teams": ["BE"],
+                "optimization_suggestions": [
+                    {
+                        "type": "HEALTHCHECK",
+                        "description": "確認 healthcheck 指令在容器內可執行 (mc/curl 是否存在)",
+                        "kubectl_or_config": f"ssh {host} 'docker exec {container_name} sh -c \"mc ready local || curl -f http://localhost:9000/minio/health/live\"'",
+                    }
+                ],
+                "reasoning": f"[規則匹配] Docker healthcheck 失敗先 restart 恢復服務，同時確認 healthcheck 指令正確。{signoz_correlation}",
+                "deviation_analysis": "容器健康檢查連續失敗，超出允許次數",
+                "confidence": 0.0,
+                "affected_services": [container_name],
+                "signoz_correlation": signoz_correlation,
+            }
+        elif alertname == "TargetDown" or (labels.get("job") and ":" in raw_target):
+            instance = labels.get("instance", raw_target)
+            job = labels.get("job", "exporter")
+            host = instance.split(":")[0]
+            mock_response = {
+                "action_title": f"檢查 {job} ({instance}) 服務存活",
+                "description": f"⚙️ 規則匹配: Prometheus 無法抓取 {instance} ({job}) 指標。常見原因: 主機離線、exporter crash、防火牆封鎖。{signoz_summary}",
+                "suggested_action": "RESTART_DEPLOYMENT",
+                "kubectl_command": f"ssh {host} 'systemctl status node_exporter 2>/dev/null || docker ps | grep exporter'",
+                "target_resource": instance,
+                "namespace": namespace,
+                "risk_level": "medium",
+                "blast_radius": {
+                    "affected_pods": 0,
+                    "estimated_downtime": "監控盲區持續中",
+                    "related_services": [job],
+                    "data_impact": "NONE",
+                },
+                "primary_responsibility": "INFRA",
+                "responsibility_reasoning": "Prometheus scrape 目標下線屬基礎設施監控範疇",
+                "secondary_teams": [],
+                "optimization_suggestions": [
+                    {
+                        "type": "MONITORING",
+                        "description": f"確認 {host} 上的 {job} exporter 是否正常運行",
+                        "kubectl_or_config": f"ssh {host} 'ps aux | grep exporter'",
+                    }
+                ],
+                "reasoning": f"[規則匹配] Prometheus target 下線，先 SSH 確認主機存活再重啟 exporter。{signoz_correlation}",
+                "deviation_analysis": "Prometheus scrape 失敗，監控數據中斷",
+                "confidence": 0.0,
+                "affected_services": [instance],
+                "signoz_correlation": signoz_correlation,
+            }
+        elif "oom" in message.lower() or "memory" in alert_type.lower():
             mock_response = {
                 "action_title": f"刪除異常 Pod {target} (OOMKilled)",
                 "description": f"⚙️ 規則匹配: {target} 發生 OOMKilled，根因為 JVM Heap 配置與 K8s memory limit 不匹配或存在記憶體洩漏。{signoz_summary}",