diff --git a/apps/api/tests/test_p0_diagnose_routing.py b/apps/api/tests/test_p0_diagnose_routing.py
index 2b10424b..fd0cbca1 100644
--- a/apps/api/tests/test_p0_diagnose_routing.py
+++ b/apps/api/tests/test_p0_diagnose_routing.py
@@ -38,3 +38,101 @@ class TestNemotronPerTaskTimeout:
 
         assert result.success is True
         mock_nvidia.tool_call.assert_called_once()
+
+
+class TestLocalFallbackChain:
+    """require_local=True 時只走 local chain，全部失敗 → REJECT，不觸碰雲端"""
+
+    @pytest.mark.asyncio
+    async def test_require_local_skips_cloud_providers(self):
+        """require_local=True 時，cloud provider 不被呼叫"""
+        import os
+        from src.services.ai_router import AIRouterExecutor, AIProviderRegistry
+        from src.services.ai_providers.interfaces import AIResult
+
+        registry = AIProviderRegistry()
+
+        # Mock: Ollama 成功
+        mock_ollama = AsyncMock()
+        mock_ollama.name = "ollama"
+        mock_ollama.privacy_level = "local"
+        mock_ollama.is_enabled = True
+        mock_ollama.capabilities = {"rca", "chat"}
+        mock_ollama.analyze = AsyncMock(return_value=AIResult(
+            raw_response="本地診斷結果",
+            success=True,
+            provider="ollama",
+        ))
+        mock_ollama.health_check = AsyncMock(return_value=True)
+
+        # Mock: Gemini（不應該被呼叫）
+        mock_gemini = AsyncMock()
+        mock_gemini.name = "gemini"
+        mock_gemini.privacy_level = "cloud"
+        mock_gemini.is_enabled = True
+        mock_gemini.analyze = AsyncMock(return_value=AIResult(
+            raw_response="雲端結果",
+            success=True,
+            provider="gemini",
+        ))
+
+        registry._providers = {
+            "ollama": mock_ollama,
+            "gemini": mock_gemini,
+        }
+
+        executor = AIRouterExecutor(registry)
+
+        # 暫時關閉 MOCK_MODE，測試真實執行路徑
+        with patch("src.services.ai_router._settings") as mock_settings:
+                mock_settings.MOCK_MODE = False
+                result = await executor.execute(
+                    prompt="診斷這個問題",
+                    provider_order=["ollama", "gemini"],
+                    require_local=True,
+                )
+
+        assert result.success is True
+        assert result.provider == "ollama"
+        mock_gemini.analyze.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_require_local_all_fail_returns_reject(self):
+        """require_local=True 且所有 local provider 失敗 → 回傳明確錯誤"""
+        import os
+        from src.services.ai_router import AIRouterExecutor, AIProviderRegistry
+        from src.services.ai_providers.interfaces import AIResult
+
+        registry = AIProviderRegistry()
+
+        # Mock: Ollama 失敗
+        mock_ollama = AsyncMock()
+        mock_ollama.name = "ollama"
+        mock_ollama.privacy_level = "local"
+        mock_ollama.is_enabled = True
+        mock_ollama.capabilities = {"rca", "chat"}
+        mock_ollama.analyze = AsyncMock(return_value=AIResult(
+            raw_response="",
+            success=False,
+            provider="ollama",
+            error="timeout",
+        ))
+        mock_ollama.health_check = AsyncMock(return_value=False)
+
+        registry._providers = {
+            "ollama": mock_ollama,
+        }
+
+        executor = AIRouterExecutor(registry)
+
+        # 暫時關閉 MOCK_MODE + 讓 telegram import 失敗（不影響主流程）
+        with patch("src.services.ai_router._settings") as mock_settings:
+                mock_settings.MOCK_MODE = False
+                result = await executor.execute(
+                    prompt="診斷這個問題",
+                    provider_order=["ollama"],
+                    require_local=True,
+                )
+
+        assert result.success is False
+        assert result.error == "local_providers_unavailable"