feat(api): Phase 18.2 FailureWatcher 失敗自動修復閉環

2026-03-31 Claude Code (統帥批准)

新增:
- IFailureWatcher Protocol (interfaces.py)
- FailureWatcherService 失敗監聽服務
  - AI 分析失敗原因 (規則引擎 + LLM 深度分析)
  - 風險等級評估 (LOW/MEDIUM/CRITICAL)
  - LOW 風險自動修復 (Phase 18.3 實際執行)
  - MEDIUM/CRITICAL 推送 Telegram 請求授權

整合:
- executor._write_audit_log() 失敗時觸發 FailureWatcher
- 失敗分類寫入 AuditLog.failure_classification
- 自動修復結果寫入 AuditLog.auto_repair_result

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

apps/api/src/repositories/interfaces.py

@@ -435,3 +435,88 @@ class IK8sRepository(Protocol):
             True 如果 K8s 連線正常
         """
         ...
+
+
+# =============================================================================
+# Phase 18: Failure Auto-Repair Loop Protocols
+# =============================================================================
+# 2026-03-31 Claude Code (統帥批准)
+# =============================================================================
+
+
+@runtime_checkable
+class IFailureWatcher(Protocol):
+    """
+    Failure Watcher Protocol
+
+    職責: 監聽失敗事件並觸發修復流程
+    實作: FailureWatcherService
+
+    版本: v1.0
+    建立: 2026-03-31 (台北時區)
+    建立者: Claude Code (Phase 18 失敗自動修復)
+    """
+
+    async def process_failure(
+        self,
+        audit_log_id: str,
+        failure_data: dict,
+    ) -> dict:
+        """
+        處理單一失敗事件
+
+        Args:
+            audit_log_id: AuditLog ID
+            failure_data: 失敗詳情
+
+        Returns:
+            {
+                "repair_attempted": bool,
+                "repair_result": str | None,
+                "risk_level": str,  # LOW/MEDIUM/CRITICAL
+                "next_action": str,  # auto_repair/await_approval/escalate
+            }
+        """
+        ...
+
+    async def analyze_failure(
+        self,
+        error_message: str,
+        operation_type: str,
+        target_resource: str,
+    ) -> dict:
+        """
+        AI 分析失敗原因
+
+        Args:
+            error_message: 錯誤訊息
+            operation_type: 操作類型
+            target_resource: 目標資源
+
+        Returns:
+            {
+                "classification": str,  # TIMEOUT/K8S_ERROR/NETWORK_ERROR/PERMISSION_DENIED
+                "root_cause": str,
+                "suggested_repair": str,
+                "risk_level": str,
+                "confidence": float,
+            }
+        """
+        ...
+
+    async def execute_auto_repair(
+        self,
+        audit_log_id: str,
+        repair_strategy: str,
+    ) -> tuple[bool, str]:
+        """
+        執行自動修復 (僅限 LOW 風險)
+
+        Args:
+            audit_log_id: 原始失敗的 AuditLog ID
+            repair_strategy: 修復策略
+
+        Returns:
+            (success, result_message)
+        """
+        ...