diff --git a/knowledge_base.py b/knowledge_base.py
index 535f166..7f8387c 100644
--- a/knowledge_base.py
+++ b/knowledge_base.py
@@ -532,6 +532,9 @@ class KnowledgeBase:
         if len(query_stripped) < 2 or len(tokens) == 0:
             return []
 
+        # ⭐ 检测安装意图：查询包含"安装/怎么装/如何安装"等词
+        has_install_keyword = any(w in query_stripped for w in ('安装', '怎么装', '如何安装', '安装方法', '安装步骤', '安装教程'))
+
         # ⭐ 过滤查询停用词：通用名词（服务器、玩家、游戏等）会导致TF-IDF虚高
         #   修复"服务器的公告"被player.md"运营方权利"压过index.md"近期公告"的问题
         from auto_reply import AutoReply
@@ -583,6 +586,10 @@ class KnowledgeBase:
         # 核心词：查询tokens中长度>=2且不在停用词表中的词
         from auto_reply import AutoReply
         query_core_tokens = set(t for t in jieba.cut(query_stripped) if len(t) >= 2 and t not in AutoReply._STOPWORDS)
+        # ⭐ 修复：安装意图查询时，将"安装"也加入核心词列表
+        # 防止"背包插件怎么装"被protect.md的"登录插件"链接文本压过
+        if has_install_keyword:
+            query_core_tokens.add('安装')
         for i, doc in enumerate(self.docs):
             heading = doc.get("heading", "")
             # 清理链接语法
@@ -884,6 +891,15 @@ class KnowledgeBase:
                             body = doc.get("body", "").strip()
                             heading = doc.get("heading", "")
                             if body.startswith('|') and '|' in body[1:] and ('操作' in body or '支持' in body):
+                                # ⭐ 修复：安装/下载意图查询时，不强制加入无关表格
+                                # 防止"背包插件怎么装"被protect.md"操作方式|是否支持"表格压过
+                                _query_is_install_or_download = has_install_keyword or download_intent
+                                if _query_is_install_or_download:
+                                    # 安装/下载意图：表格必须含安装/下载相关词才boost
+                                    _install_or_dl_terms = {'安装', '下载', '前置', '插件', '依赖'}
+                                    _tbl_has_relevance = any(t in body for t in _install_or_dl_terms)
+                                    if not _tbl_has_relevance:
+                                        continue  # 跳过无关表格
                                 # 检查heading是否与查询语义相关
                                 h_tokens = set(t for t in jieba.cut(heading) if len(t) >= 2)
                                 heading_relevant = bool(query_tokens_set & h_tokens)