diff --git a/skills/obsidian-kb/SKILL.md b/skills/obsidian-kb/SKILL.md
index b042301..37ba5ee 100644
--- a/skills/obsidian-kb/SKILL.md
+++ b/skills/obsidian-kb/SKILL.md
@@ -1,10 +1,27 @@
 ---
 name: obsidian-kb
-description: Multi-source content knowledge base maintenance system with long video segment analysis. Use when managing Obsidian vaults, processing video content from Bilibili/YouTube/Zhihu, or generating markdown notes from videos.
+description: Multi-source content knowledge base with video processing. Auto-triggers on B站/bilibili video links (BV号, b23.tv, www.bilibili.com), YouTube videos, or Zhihu articles. Use when user mentions adding videos to inbox, processing video content, extracting subtitles, or generating markdown notes from B站/YouTube/Zhihu content.
 ---
 
 # SKILL.md - Obsidian KB Skill
 
+## 🚀 重要：处理流程
+
+**当用户发送视频链接并要求保存到知识库时，直接处理，不要问用户！**
+
+正确流程：
+1. 用户发送B站链接（BV号、完整URL、或短链接如 b23.tv/xxxxx）
+2. 用户说"保存到知识库"、"添加到知识库"、"处理这个视频"等
+3. **直接使用** `workspace/skills/obsidian-kb` 处理
+4. 不要检查全局命令，不要问用户"要安装工具还是手动处理"
+
+**示例触发词：**
+- "保存到知识库"
+- "添加到知识库"
+- "处理这个视频"
+- "提取字幕生成笔记"
+- "生成obsidian笔记"
+
 ## 元数据
 
 - **名称**: obsidian-kb
diff --git a/skills/obsidian-kb/src/extractors/bilibili_extractor.py b/skills/obsidian-kb/src/extractors/bilibili_extractor.py
index d0eade3..0472250 100644
--- a/skills/obsidian-kb/src/extractors/bilibili_extractor.py
+++ b/skills/obsidian-kb/src/extractors/bilibili_extractor.py
@@ -92,6 +92,15 @@ class BilibiliExtractor(Extractor):
 
         # 解析并获取BV号
         bvid = self._resolve_bvid(source_id)
+
+        # 如果解析失败，检查是否为短链接
+        if not bvid and self.resolver.is_bilibili_short_link(source_id):
+            # 使用Playwright解析短链接
+            try:
+                bvid = asyncio.run(self._resolve_shortlink_bvid(source_id))
+            except Exception as e:
+                raise ExtractionError(f"短链接解析失败: {e}", source_id=source_id)
+
         if not bvid:
             raise ExtractionError(f"无法解析BV号: {source_id}", source_id=source_id)
 
@@ -123,13 +132,66 @@ class BilibiliExtractor(Extractor):
         if url_match:
             return url_match.group(1)
 
-        # 3. 检查是否为短链接
-        if self.resolver.is_bilibili_short_link(source_id):
-            # 短链接需要Playwright解析，这里返回None让调用方处理
-            # 或者可以在这里尝试解析，但这会使得方法变复杂
-            # 为了简化，我们假设用户会提供完整URL或BV号
-            # 短链接解析可以在extract方法中通过Playwright完成
-            pass
+        # 3. 短链接处理已移至 extract() 方法中的 _resolve_shortlink_bvid()
+        # 这里不再处理，避免混淆
+
+        return None
+
+    async def _resolve_shortlink_bvid(self, shortlink: str) -> Optional[str]:
+        """
+        使用 Playwright 解析B站短链接，获取 BV 号
+
+        Args:
+            shortlink: B站短链接（如 https://b23.tv/xxxxx）
+
+        Returns:
+            BV号字符串，解析失败返回None
+        """
+        async with async_playwright() as playwright:
+            browser = None
+            try:
+                # 启动浏览器（stealth 模式）
+                browser = await playwright.chromium.launch(
+                    headless=self.headless,
+                    args=[
+                        '--no-sandbox',
+                        '--disable-setuid-sandbox',
+                        '--disable-blink-features=AutomationControlled',
+                        '--disable-dev-shm-usage',
+                        '--start-maximized',
+                    ]
+                )
+
+                context = await browser.new_context(
+                    viewport={'width': 1920, 'height': 1080},
+                    user_agent='Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
+                    locale='zh-CN',
+                    timezone_id='Asia/Shanghai',
+                )
+
+                # 应用 stealth 模式（如果可用）
+                if STEALTH_AVAILABLE:
+                    await stealth_async(context)
+
+                page = await context.new_page()
+
+                # 访问短链接（等待重定向完成）
+                await page.goto(shortlink, wait_until='domcontentloaded', timeout=self.timeout)
+                await asyncio.sleep(3)  # 等待重定向完成
+
+                # 获取重定向后的 URL
+                final_url = page.url
+
+                # 从 URL 中提取 BV 号
+                bv_match = re.search(r'BV[a-zA-Z0-9]{10}', final_url)
+                if bv_match:
+                    return bv_match.group(0)
+
+            except Exception as e:
+                raise ExtractionError(f"短链接解析异常: {e}", source_id=shortlink)
+            finally:
+                if browser:
+                    await browser.close()
 
         return None