UPDATE

API_RELIABILITY_AUDIT_FA.md

@@ -489,25 +489,6 @@
 - سطح اعتماد:
   - `کم تا متوسط`
 
-### `POST /api/rag/recommend/irrigation/`
-
-- ماهیت:
-  - عملا همان recommendation RAG-based آبیاری را مستقیما expose می‌کند.
-- ضعف‌ها:
-  - همان ضعف‌های `POST /api/irrigation/recommend/` را دارد.
-  - ظاهر کامل پاسخ می‌تواند fallback بودن بخش‌هایی از محتوا را پنهان کند، هرچند provenance اضافه شده است.
-- سطح اعتماد:
-  - `متوسط` با ورودی خوب
-  - `کم تا متوسط` با fallback زیاد
-
-### `POST /api/rag/recommend/fertilization/`
-
-- ماهیت:
-  - مستقیم‌ترین مسیر recommendation کودهی مبتنی بر RAG است.
-- ضعف‌ها:
-  - همان ضعف‌های `POST /api/fertilization/recommend/` را دارد.
-- سطح اعتماد:
-  - `متوسط رو به کم`
 
 ---
 
@@ -553,4 +534,3 @@
 4. افزودن uncertainty و freshness صریح به heatmap و water-need outputs
 5. crop-specific و region-specific کردن alert/risk thresholds
 6. اضافه کردن confidence واقعی و source provenance استاندارد به تمام endpointهای RAG-based
-

APPS_URLS_AUDIT.md

@@ -30,8 +30,6 @@ Base: `/api/rag/`
 | Method | URL | توضیح |
 |---|---|---|
 | POST | `/api/rag/chat/` | چت RAG به صورت stream |
-| POST | `/api/rag/recommend/irrigation/` | توصیه آبیاری |
-| POST | `/api/rag/recommend/fertilization/` | توصیه کودهی |
 
 ### App: Farm Alerts
 
@@ -354,4 +352,3 @@ Base: `/api/crop-simulation/`
 - `rag/services/irrigation.py:147`
 - `rag/services/fertilization.py:130`
 - `crop_simulation/growth_simulation.py:404`
-

config/settings.py

@@ -152,7 +152,6 @@ SPECTACULAR_SETTINGS = {
         {"name": "Dashboard Data", "description": "تجمیع داده‌های داشبورد مزرعه"},
         {"name": "Farm Alerts", "description": "tracker و timeline مستقل هشدارهای مزرعه"},
         {"name": "RAG Chat", "description": "چت هوشمند RAG"},
-        {"name": "RAG Recommendations", "description": "توصیه‌های آبیاری و کودهی مبتنی بر RAG"},
         {"name": "Soil Data", "description": "داده‌های خاک (SoilGrids)"},
         {"name": "Soile", "description": "heatmap مستقل رطوبت خاک و داده های مزرعه"},
         {"name": "Farm Data", "description": "داده‌های مزرعه و سنسورها"},

config/tones/chat_tone.txt

@@ -1,14 +1,51 @@
-شما دستيار عمومي CropLogic براي چت با کاربر هستيد.
+You are a general farm assistant for CropLogic.
 
-قواعد مهم:
-- اين سرويس خروجی را به صورت متن استريمي `text/plain` برمي گرداند، نه JSON.
-- بنابراين فقط متن ساده و خوانا توليد کن و هرگز JSON، markdown fence يا ساختار کدي برنگردان.
-- پاسخ را به فارسي، دوستانه، شفاف و کاربردي بنويس.
-- اگر لازم بود، پاسخ را در 2 تا 4 پاراگراف کوتاه يا چند خط فهرست گونه اما بدون JSON ارائه کن.
-- اگر داده کافي نيست، همان را صريح بگو و از حدس زدن پرهيز کن.
+### GOAL
+Convert farm context, soil data, plant stage, weather risk, and any optimization block such as `[خروجي بهينه ساز شبيه سازي]` into a precise and practical Persian response for the farmer.
 
-شکل خروجي مورد انتظار:
-- يک پاسخ متني يکپارچه
-- بدون کليد JSON
-- بدون `sections`
-- بدون کاراکترهاي ابتدايي/انتهايي اضافه
+### HARD RULES
+1. If an optimizer block exists, it is the source of truth for numeric recommendations, timing, validity period, and scientific reasoning.
+2. Do not invent numbers or recommendations that conflict with the optimizer block.
+3. Always return only valid JSON with a top-level `sections` array.
+4. The `sections` array must include at least:
+   - one `recommendation` section for the core answer or action
+   - one `list` section for operational notes, follow-up checks, or execution details
+   - one `warning` section when there is uncertainty, data limitation, weather risk, sensor conflict, or execution risk
+5. Write in clear Persian and keep the answer practical, short, and field-usable.
+
+### OUTPUT CONTRACT
+{
+  "sections": [
+    {
+      "type": "recommendation",
+      "title": "جمع بندي اصلي",
+      "icon": "message-circle",
+      "content": "خلاصه يک جمله اي از بهترين پاسخ يا اقدام اصلي",
+      "primaryAction": "اقدام اصلي پيشنهادي",
+      "timing": "بهترين زمان اجرا يا بررسي",
+      "validityPeriod": "مدت اعتبار اين پاسخ يا توصيه",
+      "expandableExplanation": "توضيح روشن درباره دليل پاسخ با ارجاع به داده مزرعه، آب و هوا، گياه و شبيه سازي"
+    },
+    {
+      "type": "list",
+      "title": "نکات اجرايي",
+      "icon": "list",
+      "items": [
+        "نکته عملي 1",
+        "نکته عملي 2"
+      ]
+    },
+    {
+      "type": "warning",
+      "title": "هشدار يا محدوديت",
+      "icon": "alert-triangle",
+      "content": "هشدار کوتاه و کاربردي"
+    }
+  ]
+}
+
+### WRITING RULES
+- If the user asks a general question, still shape the answer inside the same `sections` contract.
+- If the optimizer highlights an important tradeoff or dominant issue, explain it briefly in `expandableExplanation`.
+- If data is incomplete or conflicting, state that clearly in the `warning` section.
+- Never output markdown, code fences, greetings, or extra commentary outside the JSON object.

docker-compose-prod.yaml

@@ -16,14 +16,14 @@ services:
       timeout: 5s
       retries: 5
     networks:
-      - ai-network
+      - crop_network
 
   redis:
     image: docker.iranserver.com/redis
     container_name: ai-redis
     restart: always
     networks:
-      - ai-network
+      - crop_network
 
   qdrant:
     image: docker.iranserver.com/qdrant/qdrant:latest
@@ -32,7 +32,7 @@ services:
     volumes:
       - qdrant_data:/qdrant/storage
     networks:
-      - ai-network
+      - crop_network
 
   web:
     build:
@@ -50,10 +50,10 @@ services:
     env_file:
       - .env
     environment:
-      DB_HOST: db
-      CELERY_BROKER_URL: redis://redis:6379/0
-      CELERY_RESULT_BACKEND: redis://redis:6379/0
-      QDRANT_HOST: qdrant
+      DB_HOST: ai-db
+      CELERY_BROKER_URL: redis://ai-redis:6379/0
+      CELERY_RESULT_BACKEND: redis://ai-redis:6379/0
+      QDRANT_HOST: ai-qdrant
       QDRANT_PORT: 6333
       DEBUG: "False"
     depends_on:
@@ -63,7 +63,7 @@ services:
         condition: service_started
 
     networks:
-      - ai-network
+      - crop_network
 
   celery:
     build:
@@ -83,9 +83,9 @@ services:
     env_file:
       - .env
     environment:
-      DB_HOST: db
-      CELERY_BROKER_URL: redis://redis:6379/0
-      CELERY_RESULT_BACKEND: redis://redis:6379/0
+      DB_HOST: ai-db
+      CELERY_BROKER_URL: redis://ai-redis:6379/0
+      CELERY_RESULT_BACKEND: redis://ai-redis:6379/0
       SKIP_MIGRATE: "1"
       DEBUG: "False"
     depends_on:
@@ -94,7 +94,7 @@ services:
       redis:
         condition: service_started
     networks:
-      - ai-network
+      - crop_network
 
 
 volumes:
@@ -102,5 +102,5 @@ volumes:
   qdrant_data:
 
 networks:
-  ai-network:
-    driver: bridge
+  crop_network:
+    external: true

docker-compose.yaml

@@ -2,7 +2,7 @@
 services:
   db:
     image:  docker.iranserver.com/mysql:8
-    container_name: ai-db
+    container_name: ai-mysql
     environment:
       MYSQL_DATABASE: ${DB_NAME:-ai}
       MYSQL_USER: ${DB_USER:-ai}
@@ -10,17 +10,21 @@ services:
       MYSQL_ROOT_PASSWORD: ${DB_PASSWORD:-changeme}
     volumes:
       - ai_mysql_data:/var/lib/mysql
+    ports:
+      - "3307:3306"
     healthcheck:
       test: ["CMD", "mysqladmin", "ping", "-h", "localhost", "-u", "root", "-p${DB_PASSWORD:-changeme}"]
       interval: 5s
       timeout: 5s
       retries: 5
+    networks:
+      - crop_network
 
   phpmyadmin:
     image: docker-mirror.liara.ir/phpmyadmin:latest
     container_name: ai-phpmyadmin
     environment:
-      PMA_HOST: db
+      PMA_HOST: ai-mysql
       PMA_PORT: 3306
       UPLOAD_LIMIT: 64M
     ports:
@@ -28,10 +32,14 @@ services:
     depends_on:
       db:
         condition: service_healthy
+    networks:
+      - crop_network
 
   redis:
     image: redis:7-alpine
     container_name: ai-redis
+    networks:
+      - crop_network
 
   qdrant:
     image: qdrant/qdrant:latest
@@ -42,6 +50,8 @@ services:
     volumes:
       - qdrant_data:/qdrant/storage
     restart: unless-stopped
+    networks:
+      - crop_network
 
   web:
     build:
@@ -61,10 +71,10 @@ services:
     env_file:
       - .env
     environment:
-      DB_HOST: db
-      CELERY_BROKER_URL: redis://redis:6379/0
-      CELERY_RESULT_BACKEND: redis://redis:6379/0
-      QDRANT_HOST: qdrant
+      DB_HOST: ai-mysql
+      CELERY_BROKER_URL: redis://ai-redis:6379/0
+      CELERY_RESULT_BACKEND: redis://ai-redis:6379/0
+      QDRANT_HOST: ai-qdrant
       QDRANT_PORT: 6333
     depends_on:
       db:
@@ -73,6 +83,8 @@ services:
         condition: service_started
       qdrant:
         condition: service_started
+    networks:
+      - crop_network
 
   celery:
     build:
@@ -90,16 +102,22 @@ services:
     env_file:
       - .env
     environment:
-      DB_HOST: db
-      CELERY_BROKER_URL: redis://redis:6379/0
-      CELERY_RESULT_BACKEND: redis://redis:6379/0
+      DB_HOST: ai-mysql
+      CELERY_BROKER_URL: redis://ai-redis:6379/0
+      CELERY_RESULT_BACKEND: redis://ai-redis:6379/0
       SKIP_MIGRATE: "1"
     depends_on:
       db:
         condition: service_healthy
       redis:
         condition: service_started
+    networks:
+      - crop_network
 
 volumes:
   ai_mysql_data:
   qdrant_data:
+
+networks:
+  crop_network:
+    external: true

fertilization/views.py

@@ -99,7 +99,9 @@ class FertilizationRecommendView(APIView):
                 status=status.HTTP_500_INTERNAL_SERVER_ERROR,
             )
 
+        # Public API exposes only the final farmer-facing recommendation object.
+        final_result = {"sections": result.get("sections", [])}
         return Response(
-            {"code": 200, "msg": "success", "data": result},
+            {"code": 200, "msg": "success", "data": final_result},
             status=status.HTTP_200_OK,
         )

integration_tests/test_reporting_and_ai_api_flow.py

@@ -239,14 +239,13 @@ class ReportingAndAiJourneyTests(IntegrationAPITestCase):
                 patch(
                     "rag.services.irrigation.get_irrigation_recommendation",
                     return_value={
-                        "plan": {
-                            "frequencyPerWeek": 3,
-                            "durationMinutes": 28,
-                            "bestTimeOfDay": "05:30",
-                            "moistureLevel": 68,
-                            "warning": "",
-                        },
-                        "raw_response": "{\"plan\": {\"frequencyPerWeek\": 3}}",
+                        "sections": [
+                            {
+                                "type": "recommendation",
+                                "title": "برنامه آبیاری بهینه",
+                                "content": "هفته ای 3 نوبت آبیاری انجام شود.",
+                            }
+                        ],
                     },
                 )
             )
@@ -254,11 +253,14 @@ class ReportingAndAiJourneyTests(IntegrationAPITestCase):
                 patch(
                     "rag.services.fertilization.get_fertilization_recommendation",
                     return_value={
-                        "plan": {
-                            "npkRatio": "15-5-30",
-                            "amountPerHectare": "60 kg",
-                        },
-                        "raw_response": "{\"plan\": {\"npkRatio\": \"15-5-30\"}}",
+                        "sections": [
+                            {
+                                "type": "recommendation",
+                                "title": "برنامه کودهی بهینه",
+                                "fertilizerType": "15-5-30",
+                                "amount": "60 kg",
+                            }
+                        ],
                     },
                 )
             )
@@ -304,37 +306,6 @@ class ReportingAndAiJourneyTests(IntegrationAPITestCase):
             streamed_text = b"".join(chat_response.streaming_content).decode("utf-8")
             self.assertIn("Moisture is acceptable", streamed_text)
 
-            rag_irrigation_response = self.client.post(
-                "/api/rag/recommend/irrigation/",
-                data={
-                    "farm_uuid": str(self.farm_uuid),
-                    "plant_name": "Tomato",
-                    "growth_stage": "flowering",
-                    "irrigation_method_name": "Analytics Drip",
-                },
-                format="json",
-            )
-            self.assertEqual(rag_irrigation_response.status_code, 200)
-            self.assertEqual(
-                rag_irrigation_response.json()["data"]["plan"]["frequencyPerWeek"],
-                3,
-            )
-
-            rag_fertilization_response = self.client.post(
-                "/api/rag/recommend/fertilization/",
-                data={
-                    "farm_uuid": str(self.farm_uuid),
-                    "plant_name": "Tomato",
-                    "growth_stage": "flowering",
-                },
-                format="json",
-            )
-            self.assertEqual(rag_fertilization_response.status_code, 200)
-            self.assertEqual(
-                rag_fertilization_response.json()["data"]["plan"]["npkRatio"],
-                "15-5-30",
-            )
-
             irrigation_recommend_response = self.client.post(
                 "/api/irrigation/recommend/",
                 data={
@@ -346,6 +317,10 @@ class ReportingAndAiJourneyTests(IntegrationAPITestCase):
                 format="json",
             )
             self.assertEqual(irrigation_recommend_response.status_code, 200)
+            self.assertEqual(
+                irrigation_recommend_response.json()["data"]["sections"][0]["type"],
+                "recommendation",
+            )
 
             fertilization_recommend_response = self.client.post(
                 "/api/fertilization/recommend/",
@@ -357,6 +332,10 @@ class ReportingAndAiJourneyTests(IntegrationAPITestCase):
                 format="json",
             )
             self.assertEqual(fertilization_recommend_response.status_code, 200)
+            self.assertEqual(
+                fertilization_recommend_response.json()["data"]["sections"][0]["fertilizerType"],
+                "15-5-30",
+            )
 
             pest_detect_response = self.client.post(
                 "/api/pest-disease/detect/",

irrigation/views.py

@@ -187,8 +187,10 @@ class IrrigationRecommendView(APIView):
                 status=status.HTTP_500_INTERNAL_SERVER_ERROR,
             )
 
+        # Public API exposes only the final farmer-facing recommendation object.
+        final_result = {"sections": result.get("sections", [])}
         return Response(
-            {"code": 200, "msg": "success", "data": result},
+            {"code": 200, "msg": "success", "data": final_result},
             status=status.HTTP_200_OK,
         )
 

rag/tests/test_recommendation_api.py

@@ -1,77 +0,0 @@
-from unittest.mock import patch
-
-from django.test import TestCase, override_settings
-from rest_framework.test import APIClient
-
-
-@override_settings(ROOT_URLCONF="config.test_urls")
-class RagRecommendationApiTests(TestCase):
-    def setUp(self):
-        self.client = APIClient()
-
-    @patch("rag.services.irrigation.get_irrigation_recommendation")
-    def test_irrigation_recommendation_returns_direct_result(self, mock_get_irrigation_recommendation):
-        mock_get_irrigation_recommendation.return_value = {
-            "plan": {
-                "frequencyPerWeek": 3,
-                "durationMinutes": 25,
-            },
-            "raw_response": "{\"plan\": {\"frequencyPerWeek\": 3, \"durationMinutes\": 25}}",
-        }
-
-        response = self.client.post(
-            "/api/rag/recommend/irrigation/",
-            data={
-                "farm_uuid": "sensor-123",
-                "plant_name": "گوجه‌فرنگی",
-                "growth_stage": "میوه‌دهی",
-                "irrigation_method_name": "قطره‌ای",
-            },
-            format="json",
-        )
-
-        self.assertEqual(response.status_code, 200)
-        self.assertEqual(response.json()["data"]["plan"]["frequencyPerWeek"], 3)
-        mock_get_irrigation_recommendation.assert_called_once_with(
-            farm_uuid="sensor-123",
-            plant_name="گوجه‌فرنگی",
-            growth_stage="میوه‌دهی",
-            irrigation_method_name="قطره‌ای",
-            query=None,
-        )
-
-    @patch("rag.services.fertilization.get_fertilization_recommendation")
-    def test_fertilization_recommendation_returns_direct_result(self, mock_get_fertilization_recommendation):
-        mock_get_fertilization_recommendation.return_value = {
-            "plan": {
-                "npkRatio": "20-20-20",
-                "amountPerHectare": "80 kg",
-            },
-            "raw_response": "{\"plan\": {\"npkRatio\": \"20-20-20\", \"amountPerHectare\": \"80 kg\"}}",
-        }
-
-        response = self.client.post(
-            "/api/rag/recommend/fertilization/",
-            data={
-                "farm_uuid": "sensor-456",
-                "plant_name": "گندم",
-                "growth_stage": "رویشی",
-            },
-            format="json",
-        )
-
-        self.assertEqual(response.status_code, 200)
-        self.assertEqual(response.json()["data"]["plan"]["npkRatio"], "20-20-20")
-        mock_get_fertilization_recommendation.assert_called_once_with(
-            farm_uuid="sensor-456",
-            plant_name="گندم",
-            growth_stage="رویشی",
-            query=None,
-        )
-
-    def test_removed_status_endpoints_return_404(self):
-        irrigation_response = self.client.get("/api/rag/recommend/irrigation/sample-task/status/")
-        fertilization_response = self.client.get("/api/rag/recommend/fertilization/sample-task/status/")
-
-        self.assertEqual(irrigation_response.status_code, 404)
-        self.assertEqual(fertilization_response.status_code, 404)

rag/urls.py

@@ -2,12 +2,8 @@ from django.urls import path
 
 from .views import (
     ChatView,
-    IrrigationRecommendationView,
-    FertilizationRecommendationView,
 )
 
 urlpatterns = [
     path("chat/", ChatView.as_view()),
-    path("recommend/irrigation/", IrrigationRecommendationView.as_view(), name="recommend-irrigation"),
-    path("recommend/fertilization/", FertilizationRecommendationView.as_view(), name="recommend-fertilization"),
 ]

rag/views.py

@@ -38,14 +38,6 @@ RagValidationErrorResponseSerializer = build_envelope_serializer(
     data_required=False,
     allow_null=True,
 )
-RagIrrigationResponseSerializer = build_envelope_serializer(
-    "RagIrrigationResponseSerializer",
-    drf_serializers.JSONField(),
-)
-RagFertilizationResponseSerializer = build_envelope_serializer(
-    "RagFertilizationResponseSerializer",
-    drf_serializers.JSONField(),
-)
 
 
 class ChatView(APIView):
@@ -211,166 +203,3 @@ class ChatView(APIView):
             generate(),
             content_type="text/plain; charset=utf-8",
         )
-
-
-class IrrigationRecommendationView(APIView):
-    """
-    توصیه آبیاری به صورت مستقیم.
-    POST با farm_uuid، plant_name، growth_stage، irrigation_method_name.
-    نتیجه همان لحظه برگشت داده می‌شود.
-    """
-
-    @extend_schema(
-        tags=["RAG Recommendations"],
-        summary="درخواست توصیه آبیاری",
-        description=(
-            "داده‌های سنسور، گیاه و روش آبیاری را دریافت کرده و "
-            "توصیه آبیاری را به صورت مستقیم برمی‌گرداند."
-        ),
-        request=inline_serializer(
-            name="IrrigationRecommendationRequest",
-            fields={
-                "farm_uuid": drf_serializers.CharField(help_text="شناسه یکتای مزرعه (اجباری)"),
-                "sensor_uuid": drf_serializers.CharField(required=False, help_text="نام قدیمی برای farm_uuid"),
-                "plant_name": drf_serializers.CharField(required=False, help_text="نام گیاه"),
-                "growth_stage": drf_serializers.CharField(required=False, help_text="مرحله رشد گیاه"),
-                "irrigation_method_name": drf_serializers.CharField(required=False, help_text="نام روش آبیاری"),
-                "query": drf_serializers.CharField(required=False, help_text="سوال اختیاری"),
-            },
-        ),
-        responses={
-            200: build_response(
-                RagIrrigationResponseSerializer,
-                "توصیه آبیاری با موفقیت تولید شد.",
-            ),
-            400: build_response(
-                RagValidationErrorResponseSerializer,
-                "پارامتر ورودی نامعتبر است.",
-            ),
-            500: build_response(
-                RagValidationErrorResponseSerializer,
-                "خطا در تولید توصیه آبیاری.",
-            ),
-        },
-        examples=[
-            OpenApiExample(
-                "نمونه درخواست",
-                value={
-                    "farm_uuid": "11111111-1111-1111-1111-111111111111",
-                    "plant_name": "گوجه‌فرنگی",
-                    "growth_stage": "میوه‌دهی",
-                    "irrigation_method_name": "آبیاری قطره‌ای",
-                },
-                request_only=True,
-            ),
-        ],
-    )
-    def post(self, request: Request):
-        from rag.services.irrigation import get_irrigation_recommendation
-
-        farm_uuid = request.data.get("farm_uuid") or request.data.get("sensor_uuid")
-        if not farm_uuid:
-            return Response(
-                {"code": 400, "msg": "پارامتر farm_uuid الزامی است.", "data": None},
-                status=status.HTTP_400_BAD_REQUEST,
-            )
-
-        try:
-            result = get_irrigation_recommendation(
-                farm_uuid=str(farm_uuid),
-                plant_name=request.data.get("plant_name"),
-                growth_stage=request.data.get("growth_stage"),
-                irrigation_method_name=request.data.get("irrigation_method_name"),
-                query=request.data.get("query"),
-            )
-        except Exception:
-            logger.exception("Direct irrigation recommendation failed for farm %s", farm_uuid)
-            return Response(
-                {"code": 500, "msg": "خطا در تولید توصیه آبیاری.", "data": None},
-                status=status.HTTP_500_INTERNAL_SERVER_ERROR,
-            )
-
-        return Response(
-            {"code": 200, "msg": "success", "data": result},
-            status=status.HTTP_200_OK,
-        )
-
-
-class FertilizationRecommendationView(APIView):
-    """
-    توصیه کودهی به صورت مستقیم.
-    POST با farm_uuid، plant_name، growth_stage.
-    نتیجه همان لحظه برگشت داده می‌شود.
-    """
-
-    @extend_schema(
-        tags=["RAG Recommendations"],
-        summary="درخواست توصیه کودهی",
-        description=(
-            "داده‌های سنسور و گیاه را دریافت کرده و "
-            "توصیه کودهی را به صورت مستقیم برمی‌گرداند."
-        ),
-        request=inline_serializer(
-            name="FertilizationRecommendationRequest",
-            fields={
-                "farm_uuid": drf_serializers.CharField(help_text="شناسه یکتای مزرعه (اجباری)"),
-                "sensor_uuid": drf_serializers.CharField(required=False, help_text="نام قدیمی برای farm_uuid"),
-                "plant_name": drf_serializers.CharField(required=False, help_text="نام گیاه"),
-                "growth_stage": drf_serializers.CharField(required=False, help_text="مرحله رشد گیاه"),
-                "query": drf_serializers.CharField(required=False, help_text="سوال اختیاری"),
-            },
-        ),
-        responses={
-            200: build_response(
-                RagFertilizationResponseSerializer,
-                "توصیه کودهی با موفقیت تولید شد.",
-            ),
-            400: build_response(
-                RagValidationErrorResponseSerializer,
-                "پارامتر ورودی نامعتبر است.",
-            ),
-            500: build_response(
-                RagValidationErrorResponseSerializer,
-                "خطا در تولید توصیه کودهی.",
-            ),
-        },
-        examples=[
-            OpenApiExample(
-                "نمونه درخواست",
-                value={
-                    "farm_uuid": "11111111-1111-1111-1111-111111111111",
-                    "plant_name": "گوجه‌فرنگی",
-                    "growth_stage": "رویشی",
-                },
-                request_only=True,
-            ),
-        ],
-    )
-    def post(self, request: Request):
-        from rag.services.fertilization import get_fertilization_recommendation
-
-        farm_uuid = request.data.get("farm_uuid") or request.data.get("sensor_uuid")
-        if not farm_uuid:
-            return Response(
-                {"code": 400, "msg": "پارامتر farm_uuid الزامی است.", "data": None},
-                status=status.HTTP_400_BAD_REQUEST,
-            )
-
-        try:
-            result = get_fertilization_recommendation(
-                farm_uuid=str(farm_uuid),
-                plant_name=request.data.get("plant_name"),
-                growth_stage=request.data.get("growth_stage"),
-                query=request.data.get("query"),
-            )
-        except Exception:
-            logger.exception("Direct fertilization recommendation failed for farm %s", farm_uuid)
-            return Response(
-                {"code": 500, "msg": "خطا در تولید توصیه کودهی.", "data": None},
-                status=status.HTTP_500_INTERNAL_SERVER_ERROR,
-            )
-
-        return Response(
-            {"code": 200, "msg": "success", "data": result},
-            status=status.HTTP_200_OK,
-        )