diff --git a/api/app/core/memory/agent/utils/get_dialogs.py b/api/app/core/memory/agent/utils/get_dialogs.py
index 22555fff..ea44d0a5 100644
--- a/api/app/core/memory/agent/utils/get_dialogs.py
+++ b/api/app/core/memory/agent/utils/get_dialogs.py
@@ -82,7 +82,9 @@ async def get_chunked_dialogs(
                         pruning_config = PruningConfig(
                             pruning_switch=memory_config.pruning_enabled,
                             pruning_scene=memory_config.pruning_scene or "education",
-                            pruning_threshold=memory_config.pruning_threshold
+                            pruning_threshold=memory_config.pruning_threshold,
+                            scene_id=str(memory_config.scene_id) if memory_config.scene_id else None,
+                            ontology_classes=memory_config.ontology_classes,
                         )
                         logger.info(f"[剪枝] 加载配置: switch={pruning_config.pruning_switch}, scene={pruning_config.pruning_scene}, threshold={pruning_config.pruning_threshold}")
                         
diff --git a/api/app/core/memory/models/config_models.py b/api/app/core/memory/models/config_models.py
index ca1780aa..c2d62ac1 100644
--- a/api/app/core/memory/models/config_models.py
+++ b/api/app/core/memory/models/config_models.py
@@ -10,7 +10,7 @@ Classes:
     TemporalSearchParams: Parameters for temporal search queries
 """
 
-from typing import Optional
+from typing import Optional, List
 from pydantic import BaseModel, Field
 
 
@@ -55,17 +55,26 @@ class PruningConfig(BaseModel):
 
     Attributes:
         pruning_switch: Enable or disable semantic pruning
-        pruning_scene: Scene type for pruning ('education', 'online_service', 'outbound')
+        pruning_scene: Scene name for pruning, either a built-in key
+            ('education', 'online_service', 'outbound') or a custom scene_name
+            from ontology_scene table
         pruning_threshold: Pruning ratio (0-0.9, max 0.9 to avoid complete removal)
+        scene_id: Optional ontology scene UUID, used to load custom ontology classes
+        ontology_classes: List of class_name strings from ontology_class table,
+            injected into the prompt when pruning_scene is not a built-in scene
     """
     pruning_switch: bool = Field(False, description="Enable semantic pruning when True.")
     pruning_scene: str = Field(
         "education",
-        description="Scene for pruning: one of 'education', 'online_service', 'outbound'.",
+        description="Scene for pruning: built-in key or custom scene_name from ontology_scene.",
     )
     pruning_threshold: float = Field(
         0.5, ge=0.0, le=0.9,
         description="Pruning ratio within 0-0.9 (max 0.9 to avoid termination).")
+    scene_id: Optional[str] = Field(None, description="Ontology scene UUID (optional).")
+    ontology_classes: Optional[List[str]] = Field(
+        None, description="Class names from ontology_class table for custom scenes."
+    )
 
 
 class TemporalSearchParams(BaseModel):
diff --git a/api/app/core/memory/storage_services/extraction_engine/data_preprocessing/data_pruning.py b/api/app/core/memory/storage_services/extraction_engine/data_preprocessing/data_pruning.py
index 0a913633..904b238f 100644
--- a/api/app/core/memory/storage_services/extraction_engine/data_preprocessing/data_pruning.py
+++ b/api/app/core/memory/storage_services/extraction_engine/data_preprocessing/data_pruning.py
@@ -86,19 +86,26 @@ class SemanticPruner:
         self._detailed_prune_logging = True  # 是否启用详细日志
         self._max_debug_msgs_per_dialog = 20  # 每个对话最多记录前N条消息的详细日志
         
-        # 加载场景特定配置
+        # 加载场景特定配置（内置场景走专门规则，自定义场景 fallback 到通用规则）
         self.scene_config: ScenePatterns = SceneConfigRegistry.get_config(
             self.config.pruning_scene, 
             fallback_to_generic=True
         )
         
-        # 检查场景是否有专门支持
-        is_supported = SceneConfigRegistry.is_scene_supported(self.config.pruning_scene)
-        if is_supported:
-            self._log(f"[剪枝-初始化] 场景={self.config.pruning_scene} 使用专门配置")
+        # 判断是否为内置专门场景
+        self._is_builtin_scene = SceneConfigRegistry.is_scene_supported(self.config.pruning_scene)
+        
+        # 自定义场景的本体类型列表（用于注入提示词）
+        self._ontology_classes = getattr(self.config, "ontology_classes", None) or []
+        
+        if self._is_builtin_scene:
+            self._log(f"[剪枝-初始化] 场景={self.config.pruning_scene} 使用内置专门配置")
         else:
-            self._log(f"[剪枝-初始化] 场景={self.config.pruning_scene} 未预定义，使用通用配置（保守策略）")
-            self._log(f"[剪枝-初始化] 支持的场景: {SceneConfigRegistry.get_all_scenes()}")
+            self._log(f"[剪枝-初始化] 场景={self.config.pruning_scene} 为自定义场景，使用通用规则 + 本体类型提示词注入")
+            if self._ontology_classes:
+                self._log(f"[剪枝-初始化] 注入本体类型: {self._ontology_classes}")
+            else:
+                self._log(f"[剪枝-初始化] 未找到本体类型，将使用通用提示词")
         
         # Load Jinja2 template
         self.template = prompt_env.get_template("extracat_Pruning.jinja2")
@@ -424,12 +431,16 @@ class SemanticPruner:
             self._log(f"[剪枝-缓存] LRU缓存已满，删除最旧条目")
 
         rendered = self.template.render(
-            pruning_scene=self.config.pruning_scene, 
+            pruning_scene=self.config.pruning_scene,
+            is_builtin_scene=self._is_builtin_scene,
+            ontology_classes=self._ontology_classes,
             dialog_text=dialog_text,
             language=self.language
         )
         log_template_rendering("extracat_Pruning.jinja2", {
             "pruning_scene": self.config.pruning_scene,
+            "is_builtin_scene": self._is_builtin_scene,
+            "ontology_classes_count": len(self._ontology_classes),
             "language": self.language
         })
         log_prompt_rendering("pruning-extract", rendered)
diff --git a/api/app/core/memory/utils/prompt/prompts/extracat_Pruning.jinja2 b/api/app/core/memory/utils/prompt/prompts/extracat_Pruning.jinja2
index 8253924b..6b620df9 100644
--- a/api/app/core/memory/utils/prompt/prompts/extracat_Pruning.jinja2
+++ b/api/app/core/memory/utils/prompt/prompts/extracat_Pruning.jinja2
@@ -1,6 +1,6 @@
 {#
   对话级抽取与相关性判定模板（用于剪枝加速）
-  输入：pruning_scene, dialog_text
+  输入：pruning_scene, is_builtin_scene, ontology_classes, dialog_text, language
   输出：严格 JSON（不要包含任何多余文本），字段：
     - is_related: bool，是否与所选场景相关
     - times: [string]，从对话中抽取的时间相关文本（日期、时间、时间段、有效期等）
@@ -16,7 +16,8 @@
   - 仅输出上述键；避免多余解释或字段。
 #}
 
-{% set scene_instructions = {
+{# ── 内置场景的固定说明 ── #}
+{% set builtin_scene_instructions = {
   'education': {
     'zh': '教育场景：教学、课程、考试、作业、老师/学生互动、学习资源、学校管理等。',
     'en': 'Education Scenario: Teaching, courses, exams, homework, teacher/student interaction, learning resources, school management, etc.'
@@ -31,16 +32,40 @@
   }
 } %}
 
-{% set scene_key = pruning_scene %}
-{% if scene_key not in scene_instructions %}
-{% set scene_key = 'education' %}
+{# ── 确定最终使用的场景说明 ── #}
+{% if is_builtin_scene %}
+  {# 内置专门场景：使用固定说明 #}
+  {% set scene_key = pruning_scene %}
+  {% if scene_key not in builtin_scene_instructions %}{% set scene_key = 'education' %}{% endif %}
+  {% set instruction = builtin_scene_instructions[scene_key][language] if language in ['zh', 'en'] else builtin_scene_instructions[scene_key]['zh'] %}
+  {% set custom_types_str = '' %}
+{% else %}
+  {# 自定义场景：使用场景名称 + 本体类型列表构建说明 #}
+  {% if ontology_classes and ontology_classes | length > 0 %}
+    {% if language == 'en' %}
+      {% set custom_types_str = ontology_classes | join(', ') %}
+      {% set instruction = 'Custom scene "' ~ pruning_scene ~ '": The dialogue is related to this scene if it involves any of the following entity types: ' ~ custom_types_str ~ '.' %}
+    {% else %}
+      {% set custom_types_str = ontology_classes | join('、') %}
+      {% set instruction = '自定义场景「' ~ pruning_scene ~ '」：对话涉及以下任意实体类型时视为相关：' ~ custom_types_str ~ '。' %}
+    {% endif %}
+  {% else %}
+    {# 无本体类型时退化为通用说明 #}
+    {% if language == 'en' %}
+      {% set instruction = 'Custom scene "' ~ pruning_scene ~ '": Determine whether the dialogue content is relevant to this scene based on overall context.' %}
+    {% else %}
+      {% set instruction = '自定义场景「' ~ pruning_scene ~ '」：根据对话整体内容判断是否与该场景相关。' %}
+    {% endif %}
+    {% set custom_types_str = '' %}
+  {% endif %}
 {% endif %}
 
-{% set instruction = scene_instructions[scene_key][language] if language in ['zh', 'en'] else scene_instructions[scene_key]['zh'] %}
-
 {% if language == "zh" %}
 请在下方对话全文基础上，按该场景进行一次性抽取并判定相关性：
 场景说明：{{ instruction }}
+{% if not is_builtin_scene and custom_types_str %}
+重要提示：只要对话中出现与上述实体类型（{{ custom_types_str }}）相关的内容，即判定为相关（is_related=true）。
+{% endif %}
 
 对话全文：
 """
@@ -60,6 +85,9 @@
 {% else %}
 Based on the full dialogue below, perform one-time extraction and relevance determination according to this scenario:
 Scenario Description: {{ instruction }}
+{% if not is_builtin_scene and custom_types_str %}
+Important: If the dialogue contains content related to any of the entity types above ({{ custom_types_str }}), mark it as relevant (is_related=true).
+{% endif %}
 
 Full Dialogue:
 """
diff --git a/api/app/repositories/memory_config_repository.py b/api/app/repositories/memory_config_repository.py
index 2dae51ef..22f13449 100644
--- a/api/app/repositories/memory_config_repository.py
+++ b/api/app/repositories/memory_config_repository.py
@@ -233,6 +233,7 @@ class MemoryConfigRepository:
                 config_desc=params.config_desc,
                 workspace_id=params.workspace_id,
                 scene_id=params.scene_id,
+                pruning_scene=params.pruning_scene,
                 llm_id=params.llm_id,
                 embedding_id=params.embedding_id,
                 rerank_id=params.rerank_id,
diff --git a/api/app/schemas/memory_config_schema.py b/api/app/schemas/memory_config_schema.py
index 0b63844b..0c359d70 100644
--- a/api/app/schemas/memory_config_schema.py
+++ b/api/app/schemas/memory_config_schema.py
@@ -417,6 +417,7 @@ class MemoryConfig:
     
     # Ontology scene association
     scene_id: Optional[UUID] = None
+    ontology_classes: Optional[list] = field(default=None)
     
     def __post_init__(self):
         """Validate configuration after initialization."""
diff --git a/api/app/schemas/memory_storage_schema.py b/api/app/schemas/memory_storage_schema.py
index 776d2783..e396bbf6 100644
--- a/api/app/schemas/memory_storage_schema.py
+++ b/api/app/schemas/memory_storage_schema.py
@@ -232,14 +232,15 @@ class ConfigParamsCreate(BaseModel):  # 创建配置参数模型（仅 body，
     # 本体场景关联（可选）
     scene_id: Optional[uuid.UUID] = Field(None, description="本体场景ID（UUID），关联ontology_scene表")
     
+    # 语义剪枝场景（由 service 层根据 scene_id 自动推导，值为关联场景的 scene_name，前端无需传入）
+    pruning_scene: Optional[str] = Field(None, description="语义剪枝场景，由 scene_id 对应的 scene_name 自动填充")
+    
     # 模型配置字段（可选，用于手动指定或自动填充）
     llm_id: Optional[str] = Field(None, description="LLM模型配置ID")
     embedding_id: Optional[str] = Field(None, description="嵌入模型配置ID")
     rerank_id: Optional[str] = Field(None, description="重排序模型配置ID")
     reflection_model_id: Optional[str] = Field(None, description="反思模型ID，默认与llm_id一致")
     emotion_model_id: Optional[str] = Field(None, description="情绪分析模型ID，默认与llm_id一致")
-
-
 class ConfigParamsDelete(BaseModel):  # 删除配置参数模型（请求体）
     model_config = ConfigDict(populate_by_name=True, extra="forbid")
     # config_name: str = Field("配置名称", description="配置名称（字符串）")
diff --git a/api/app/services/memory_config_service.py b/api/app/services/memory_config_service.py
index ccfd5482..00757f8c 100644
--- a/api/app/services/memory_config_service.py
+++ b/api/app/services/memory_config_service.py
@@ -107,6 +107,40 @@ def _validate_config_id(config_id, db: Session = None):
     )
 
 
+# 专门场景的内置 key 集合，直接从 SceneConfigRegistry 派生，避免重复维护
+# 使用懒加载函数避免模块级循环导入
+def _get_builtin_pruning_scenes() -> set:
+    from app.core.memory.storage_services.extraction_engine.data_preprocessing.scene_config import SceneConfigRegistry
+    return set(SceneConfigRegistry.get_all_scenes())
+
+
+def _load_ontology_classes(db: Session, scene_id, pruning_scene: Optional[str]) -> Optional[list]:
+    """当 pruning_scene 不是内置场景时，从 ontology_class 表加载类型名称列表。
+
+    Args:
+        db: 数据库会话
+        scene_id: 本体场景 UUID
+        pruning_scene: 语义剪枝场景名称
+
+    Returns:
+        class_name 字符串列表，或 None（内置场景 / 无数据时）
+    """
+    if not scene_id:
+        return None
+    # 内置场景走 SceneConfigRegistry，不需要注入类型列表
+    if pruning_scene in _get_builtin_pruning_scenes():
+        return None
+    try:
+        from app.repositories.ontology_class_repository import OntologyClassRepository
+        repo = OntologyClassRepository(db)
+        classes = repo.get_classes_by_scene(scene_id)
+        names = [c.class_name for c in classes if c.class_name]
+        return names if names else None
+    except Exception as e:
+        logger.warning(f"Failed to load ontology classes for scene_id={scene_id}: {e}")
+        return None
+
+
 class MemoryConfigService:
     """
     Centralized service for memory configuration loading and validation.
@@ -359,6 +393,7 @@ class MemoryConfigService:
                 pruning_threshold=float(memory_config.pruning_threshold) if memory_config.pruning_threshold is not None else 0.5,
                 # Ontology scene association
                 scene_id=memory_config.scene_id,
+                ontology_classes=_load_ontology_classes(self.db, memory_config.scene_id, memory_config.pruning_scene),
             )
 
             elapsed_ms = (time.time() - start_time) * 1000
diff --git a/api/app/services/memory_storage_service.py b/api/app/services/memory_storage_service.py
index 02fd1051..a83d6830 100644
--- a/api/app/services/memory_storage_service.py
+++ b/api/app/services/memory_storage_service.py
@@ -146,6 +146,10 @@ class DataConfigService: # 数据配置服务类（PostgreSQL）
         if not params.emotion_model_id:
             params.emotion_model_id = params.llm_id
 
+        # 根据关联的本体场景推导 pruning_scene（语义剪枝场景与本体工程场景保持一致）
+        if params.scene_id and not getattr(params, 'pruning_scene', None):
+            params.pruning_scene = self._resolve_pruning_scene_from_scene_id(params.scene_id)
+
         config = MemoryConfigRepository.create(self.db, params)
         self.db.commit()
         return {"affected": 1, "config_id": config.config_id}
@@ -161,6 +165,23 @@ class DataConfigService: # 数据配置服务类（PostgreSQL）
         finally:
             db_session.close()
 
+    def _resolve_pruning_scene_from_scene_id(self, scene_id) -> Optional[str]:
+        """根据本体场景ID获取对应的 scene_name，作为语义剪枝场景值
+
+        Args:
+            scene_id: 本体场景UUID
+
+        Returns:
+            scene_name 字符串，查询失败时返回 None
+        """
+        try:
+            from app.models.ontology_scene import OntologyScene
+            scene = self.db.query(OntologyScene).filter_by(scene_id=scene_id).first()
+            return scene.scene_name if scene else None
+        except Exception as e:
+            logger.warning(f"_resolve_pruning_scene_from_scene_id failed for scene_id={scene_id}: {e}", exc_info=True)
+            return None
+
     # --- Delete ---
     def delete(self, key: ConfigParamsDelete) -> Dict[str, Any]: # 删除配置参数（按配置ID）
         success = MemoryConfigRepository.delete(self.db, key.config_id)
@@ -196,6 +217,19 @@ class DataConfigService: # 数据配置服务类（PostgreSQL）
     def get_all(self, workspace_id = None) -> List[Dict[str, Any]]: # 获取所有配置参数
         results = MemoryConfigRepository.get_all(self.db, workspace_id)
 
+        # 检查并修正 pruning_scene 与 scene_name 不一致的记录
+        needs_commit = False
+        for config, scene_name in results:
+            if scene_name and config.pruning_scene != scene_name:
+                logger.info(
+                    f"修正 pruning_scene: config_id={config.config_id} "
+                    f"'{config.pruning_scene}' -> '{scene_name}'"
+                )
+                config.pruning_scene = scene_name
+                needs_commit = True
+        if needs_commit:
+            self.db.commit()
+
         # 将 ORM 对象转换为字典列表
         data_list = []
         for config, scene_name in results:
diff --git a/api/app/services/workspace_service.py b/api/app/services/workspace_service.py
index e93c0c5c..74880410 100644
--- a/api/app/services/workspace_service.py
+++ b/api/app/services/workspace_service.py
@@ -152,6 +152,7 @@ def create_workspace(
 
         # Initialize default ontology scenes for the workspace (先创建本体场景)
         default_scene_id = None
+        default_scene_name = None
         try:
             initializer = DefaultOntologyInitializer(db)
             success, error_msg = initializer.initialize_default_scenes(
@@ -163,7 +164,7 @@ def create_workspace(
                     f"为工作空间 {db_workspace.id} 创建默认本体场景成功 (language={language})"
                 )
                 
-                # 获取默认场景ID，优先使用"在线教育"场景，如果不存在则使用"情感陪伴"场景
+        # 获取默认场景ID，优先使用"在线教育"场景，如果不存在则使用"情感陪伴"场景
                 from app.repositories.ontology_scene_repository import OntologySceneRepository
                 from app.config.default_ontology_config import (
                     ONLINE_EDUCATION_SCENE, 
@@ -179,6 +180,7 @@ def create_workspace(
                 
                 if education_scene:
                     default_scene_id = education_scene.scene_id
+                    default_scene_name = education_scene.scene_name
                     business_logger.info(
                         f"获取到教育场景ID用于默认记忆配置: {default_scene_id} (scene_name={education_scene_name})"
                     )
@@ -189,6 +191,7 @@ def create_workspace(
                     
                     if companion_scene:
                         default_scene_id = companion_scene.scene_id
+                        default_scene_name = companion_scene.scene_name
                         business_logger.info(
                             f"教育场景不存在，使用情感陪伴场景ID用于默认记忆配置: {default_scene_id} (scene_name={companion_scene_name})"
                         )
@@ -219,6 +222,7 @@ def create_workspace(
                     embedding_id=embedding,
                     rerank_id=rerank,
                     scene_id=default_scene_id,  # 传入默认场景ID（优先教育场景，其次情感陪伴场景）
+                    pruning_scene_name=default_scene_name,  # 传入场景名称作为语义剪枝场景值
                 )
                 business_logger.info(
                     f"为工作空间 {db_workspace.id} 创建默认记忆配置成功 (scene_id={default_scene_id})"
@@ -1159,6 +1163,7 @@ def _create_default_memory_config(
     embedding_id: Optional[uuid.UUID] = None,
     rerank_id: Optional[uuid.UUID] = None,
     scene_id: Optional[uuid.UUID] = None,
+    pruning_scene_name: Optional[str] = None,
 ) -> None:
     """Create a default memory config for a newly created workspace.
     
@@ -1170,6 +1175,7 @@ def _create_default_memory_config(
         embedding_id: Optional embedding model ID
         rerank_id: Optional rerank model ID
         scene_id: Optional ontology scene ID (默认关联教育场景)
+        pruning_scene_name: Optional pruning scene name，取自 ontology_scene.scene_name
     """
     from app.models.memory_config_model import MemoryConfig
     
@@ -1183,7 +1189,8 @@ def _create_default_memory_config(
         llm_id=str(llm_id) if llm_id else None,
         embedding_id=str(embedding_id) if embedding_id else None,
         rerank_id=str(rerank_id) if rerank_id else None,
-        scene_id=scene_id,  # 关联本体场景ID
+        scene_id=scene_id,  # 关联本体场景ID（默认为"在线教育"场景）
+        pruning_scene=pruning_scene_name,  # 语义剪枝场景直接使用 scene_name
         state=True,  # Active by default
         is_default=True,  # Mark as workspace default
     )