diff --git a/db/message_storage.py b/db/message_storage.py
index 7d173f7..e597fec 100644
--- a/db/message_storage.py
+++ b/db/message_storage.py
@@ -133,15 +133,26 @@ class MessageStorageDB(BaseDBOperator):
 
         return result
 
-    def get_pending_mention_extract_messages(self, limit: int = 200, max_age_days: int = 7) -> List[Dict]:
+    def get_pending_mention_extract_messages(
+        self,
+        limit: int = 200,
+        window_start_minutes: int = 20,
+        window_end_minutes: int = 10,
+    ) -> List[Dict]:
         """获取待处理 @ 抽取的消息批次。
 
         筛选规则：
         1. 群消息；
         2. mentioned_user_ids 为空（表示还未处理）；
         3. message_xml 非空；
-        4. 时间在 max_age_days 窗口内，避免扫描无限历史。
+        4. 仅处理固定时间窗口（默认：10~20分钟前），降低扫描压力与热数据竞争。
         """
+        # 兜底修正窗口参数，确保窗口有效：start > end >= 0
+        start_m = max(int(window_start_minutes), 1)
+        end_m = max(int(window_end_minutes), 0)
+        if start_m <= end_m:
+            start_m = end_m + 10
+
         sql = """
         SELECT message_id, group_id, sender, message_xml, timestamp
         FROM messages
@@ -150,13 +161,19 @@ class MessageStorageDB(BaseDBOperator):
           AND (mentioned_user_ids IS NULL OR mentioned_user_ids = '')
           AND message_xml IS NOT NULL
           AND message_xml <> ''
-          AND timestamp >= DATE_SUB(NOW(), INTERVAL %s DAY)
+          AND timestamp >= DATE_SUB(NOW(), INTERVAL %s MINUTE)
+          AND timestamp < DATE_SUB(NOW(), INTERVAL %s MINUTE)
         ORDER BY timestamp ASC
         LIMIT %s
         """
-        return self.execute_query(sql, (max_age_days, limit)) or []
+        return self.execute_query(sql, (start_m, end_m, limit)) or []
 
-    def process_pending_mentions(self, batch_size: int = 200, max_age_days: int = 7) -> Dict[str, int]:
+    def process_pending_mentions(
+        self,
+        batch_size: int = 200,
+        window_start_minutes: int = 20,
+        window_end_minutes: int = 10,
+    ) -> Dict[str, int]:
         """批量处理待抽取 @ 的消息，并同步社交图数据。
 
         返回统计：
@@ -165,7 +182,11 @@ class MessageStorageDB(BaseDBOperator):
         - with_mentions: 提取到 @ 的消息条数
         - failed: 失败条数
         """
-        rows = self.get_pending_mention_extract_messages(limit=batch_size, max_age_days=max_age_days)
+        rows = self.get_pending_mention_extract_messages(
+            limit=batch_size,
+            window_start_minutes=window_start_minutes,
+            window_end_minutes=window_end_minutes,
+        )
         if not rows:
             return {"total": 0, "processed": 0, "with_mentions": 0, "failed": 0}
 
diff --git a/utils/system_jobs.py b/utils/system_jobs.py
index fe34b6c..7f04aa6 100644
--- a/utils/system_jobs.py
+++ b/utils/system_jobs.py
@@ -64,7 +64,11 @@ def _build_process_pending_images_handler(robot) -> Callable[[], Awaitable[None]
 def _build_process_pending_mentions_handler(robot) -> Callable[[], Awaitable[None]]:
     async def _handler():
         if hasattr(robot, "message_storage") and robot.message_storage:
-            await robot.message_storage.process_pending_mentions(batch_size=200, max_age_days=7)
+            await robot.message_storage.process_pending_mentions(
+                batch_size=200,
+                window_start_minutes=20,
+                window_end_minutes=10,
+            )
 
     return _handler
 
diff --git a/utils/wechat/message_to_db.py b/utils/wechat/message_to_db.py
index 92abfdd..f207fdc 100644
--- a/utils/wechat/message_to_db.py
+++ b/utils/wechat/message_to_db.py
@@ -358,18 +358,25 @@ class MessageStorage:
         except Exception as e:
             logger.exception(f"定时处理媒体任务出错: {e}")
 
-    async def process_pending_mentions(self, batch_size: int = 200, max_age_days: int = 7):
+    async def process_pending_mentions(
+        self,
+        batch_size: int = 200,
+        window_start_minutes: int = 20,
+        window_end_minutes: int = 10,
+    ):
         """定时任务：批量处理待抽取 @ 的消息并写入社交图。
 
         说明：
         1. 该任务与主消息归档链路解耦，不阻塞实时收发；
         2. 每次只处理有限批次，避免长事务和数据库抖动；
         3. 重复执行安全：底层按 message_id + sender + mentioned_user_id 做幂等控制。
+        4. 默认只处理 10~20 分钟前的数据，减少对热数据区间的扫描压力。
         """
         try:
             stats = self.message_db.process_pending_mentions(
                 batch_size=batch_size,
-                max_age_days=max_age_days,
+                window_start_minutes=window_start_minutes,
+                window_end_minutes=window_end_minutes,
             )
             total = int(stats.get("total", 0))
             if total == 0: