refactor(memory): unify user placeholder names and harden alias sync logic

- Replace hardcoded user placeholder name lists in write_tools and
user_memory_service with shared _USER_PLACEHOLDER_NAMES constant
- Filter user placeholder names during alias merging in _merge_attribute
  to prevent cross-role alias contamination on non-user entities
- Use toLower() in Cypher query for case-insensitive name matching
- Change PgSQL->Neo4j alias sync condition from 'if pg_aliases' to
  'if info is not None' so empty aliases correctly clear stale data

api/app/core/memory/agent/utils/write_tools.py

@@ -14,6 +14,7 @@ from dotenv import load_dotenv
 
 from app.core.logging_config import get_agent_logger
 from app.core.memory.agent.utils.get_dialogs import get_chunked_dialogs
+from app.core.memory.storage_services.extraction_engine.deduplication.deduped_and_disamb import _USER_PLACEHOLDER_NAMES
 from app.core.memory.storage_services.extraction_engine.extraction_orchestrator import ExtractionOrchestrator
 from app.core.memory.storage_services.extraction_engine.knowledge_extraction.memory_summary import \
     memory_summary_generation
@@ -201,14 +202,17 @@ async def write(
                             with get_db_context() as db_session:
                                 info = EndUserInfoRepository(db_session).get_by_end_user_id(uuid.UUID(end_user_id))
                                 pg_aliases = info.aliases if info and info.aliases else []
-                            if pg_aliases:
+                            if info is not None:
+                                # 将 Python 侧占位名集合作为参数传入，避免 Cypher 硬编码
+                                placeholder_names = list(_USER_PLACEHOLDER_NAMES)
                                 await neo4j_connector.execute_query(
                                     """
                                     MATCH (e:ExtractedEntity)
-                                    WHERE e.end_user_id = $end_user_id AND e.name IN ['用户', '我', 'User', 'I']
+                                    WHERE e.end_user_id = $end_user_id AND toLower(e.name) IN $placeholder_names
                                     SET e.aliases = $aliases
                                     """,
                                     end_user_id=end_user_id, aliases=pg_aliases,
+                                    placeholder_names=placeholder_names,
                                 )
                                 logger.info(f"[AliasSync] Neo4j 用户实体 aliases 已用 PgSQL 权威源覆盖: {pg_aliases}")
                     except Exception as sync_err:

api/app/core/memory/storage_services/extraction_engine/deduplication/deduped_and_disamb.py

@@ -88,11 +88,14 @@ def _merge_attribute(canonical: ExtractedEntityNode, ent: ExtractedEntityNode):
         if canonical_name.lower() not in _USER_PLACEHOLDER_NAMES:
             incoming_name = (getattr(ent, "name", "") or "").strip()
             
-            # 收集所有需要合并的别名
+            # 收集所有需要合并的别名，过滤掉用户占位名避免污染非用户实体
             all_aliases = list(getattr(canonical, "aliases", []) or [])
-            if incoming_name and incoming_name != canonical_name:
+            if incoming_name and incoming_name != canonical_name and incoming_name.lower() not in _USER_PLACEHOLDER_NAMES:
                 all_aliases.append(incoming_name)
-            all_aliases.extend(getattr(ent, "aliases", []) or [])
+            all_aliases.extend(
+                a for a in (getattr(ent, "aliases", []) or [])
+                if a and a.strip().lower() not in _USER_PLACEHOLDER_NAMES
+            )
             
             try:
                 from app.core.memory.utils.alias_utils import normalize_aliases

api/app/services/user_memory_service.py

@@ -14,6 +14,7 @@ from pydantic import BaseModel, Field
 from sqlalchemy.orm import Session
 
 from app.core.logging_config import get_logger
+from app.core.memory.storage_services.extraction_engine.deduplication.deduped_and_disamb import _USER_PLACEHOLDER_NAMES
 from app.core.memory.utils.llm.llm_utils import MemoryClientFactory
 from app.db import get_db_context
 from app.repositories.conversation_repository import ConversationRepository
@@ -473,7 +474,7 @@ class UserMemoryService:
             allowed_fields = {'other_name', 'aliases', 'meta_data'}
             
             # 用户占位名称黑名单，不允许作为 other_name 或出现在 aliases 中
-            _user_placeholder_names = {'用户', '我', 'User', 'I'}
+            _user_placeholder_names = _USER_PLACEHOLDER_NAMES
             
             # 过滤 other_name：不允许设置为占位名称
             if 'other_name' in update_data and update_data['other_name'] and update_data['other_name'].strip() in _user_placeholder_names: