feat: kb-zhihu-metadata 支持 ZHIHU_COOKIE

- 安装 markdownify 依赖
- PlaywrightBrowser 支持传入 Cookie
- ZhihuPlaywrightExtractor 支持传入 Cookie
- zhihu_metadata.py 读取 ZHIHU_COOKIE 环境变量
- 更新 SKILL.md 添加 Cookie 配置说明

kb-zhihu-metadata/SKILL.md

@@ -9,6 +9,24 @@
 - User information extraction
 - JSON output format
 - Markdown conversion support
+- **Cookie 支持**：可配置 ZHIHU_COOKIE 访问需要登录的内容
+
+## 环境变量
+
+| 变量名 | 用途 | 必需 |
+|--------|------|------|
+| `ZHIHU_COOKIE` | 知乎登录 Cookie | 推荐（访问专栏等需要） |
+
+**获取 ZHIHU_COOKIE**：
+1. 登录知乎网页版
+2. 打开开发者工具 (F12) → Application → Cookies → zhihu.com
+3. 复制 Cookie 字符串
+
+```bash
+# 添加到 ~/.kb_env
+echo 'export ZHIHU_COOKIE="你的知乎Cookie"' >> ~/.kb_env
+source ~/.kb_env
+```
 
 ## Commands
 

kb-zhihu-metadata/zhihu_extractor.py

@@ -30,16 +30,42 @@ class PlaywrightBrowser:
         self.context = None
         self.page = None
 
-    def start(self):
+    def start(self, cookie: str = None):
         """启动浏览器"""
         from playwright.sync_api import sync_playwright
         self.playwright = sync_playwright().start()
         self.browser = self.playwright.chromium.launch(headless=self.headless)
-        self.context = self.browser.new_context(
-            user_agent='Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36'
-        )
+        
+        # 创建浏览器上下文
+        context_options = {
+            'user_agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36'
+        }
+        
+        # 如果提供了 Cookie，设置 Cookie
+        if cookie:
+            context_options['cookies'] = self._parse_cookie(cookie)
+        
+        self.context = self.browser.new_context(**context_options)
         self.page = self.context.new_page()
         self.page.set_default_timeout(self.timeout)
+    
+    def _parse_cookie(self, cookie_str: str) -> list:
+        """解析 Cookie 字符串为 Playwright 格式"""
+        cookies = []
+        for item in cookie_str.split(';'):
+            item = item.strip()
+            if '=' in item:
+                name, value = item.split('=', 1)
+                cookies.append({
+                    'name': name.strip(),
+                    'value': value.strip(),
+                    'domain': '.zhihu.com',
+                    'path': '/',
+                    'secure': True,
+                    'httpOnly': False,
+                    'sameSite': 'Lax'
+                })
+        return cookies
 
     def close(self):
         """关闭浏览器"""
@@ -84,14 +110,16 @@ class ZhihuPlaywrightExtractor:
         'question': r'zhihu\.com/question/(\d+)',
     }
 
-    def __init__(self, headless: bool = True, timeout: int = 30):
+    def __init__(self, headless: bool = True, timeout: int = 30, cookie: str = None):
         """
         初始化提取器
 
         Args:
             headless: 是否使用无头浏览器
             timeout: 超时时间（秒）
+            cookie: 知乎 Cookie 字符串（可选）
         """
+        self.cookie = cookie
         self.browser = PlaywrightBrowser(headless=headless, timeout=timeout)
         self.markdown_converter = MarkdownConverter()
 
@@ -383,8 +411,8 @@ class ZhihuPlaywrightExtractor:
             print(f"URL 类型: {url_type}")
             print(f"启动浏览器...")
 
-            # 启动浏览器
-            self.browser.start()
+            # 启动浏览器（传入 Cookie）
+            self.browser.start(cookie=self.cookie)
 
             print(f"访问页面: {url}")
             # 访问页面

kb-zhihu-metadata/zhihu_metadata.py

@@ -23,43 +23,46 @@ if current_dir not in sys.path:
 from zhihu_extractor import ZhihuPlaywrightExtractor
 
 
-def extract_zvideo(video_id: str, headless: bool = True) -> Dict[str, Any]:
+def extract_zvideo(video_id: str, headless: bool = True, cookie: str = None) -> Dict[str, Any]:
     """
     获取知乎视频元数据
 
     Args:
         video_id: 视频 ID (数字字符串)
         headless: 是否使用无头浏览器
+        cookie: 知乎 Cookie（可选）
 
     Returns:
         视频元数据字典
     """
     url = f"https://www.zhihu.com/zvideo/{video_id}"
-    return _extract_metadata(url, "zvideo", headless)
+    return _extract_metadata(url, "zvideo", headless, cookie)
 
 
-def extract_zhuanlan(article_id: str, headless: bool = True) -> Dict[str, Any]:
+def extract_zhuanlan(article_id: str, headless: bool = True, cookie: str = None) -> Dict[str, Any]:
     """
     获取知乎专栏文章元数据
 
     Args:
         article_id: 文章 ID (数字字符串)
         headless: 是否使用无头浏览器
+        cookie: 知乎 Cookie（可选）
 
     Returns:
         文章元数据字典
     """
     url = f"https://zhuanlan.zhihu.com/p/{article_id}"
-    return _extract_metadata(url, "zhuanlan", headless)
+    return _extract_metadata(url, "zhuanlan", headless, cookie)
 
 
-def extract_answer(answer_id: str, headless: bool = True) -> Dict[str, Any]:
+def extract_answer(answer_id: str, headless: bool = True, cookie: str = None) -> Dict[str, Any]:
     """
     获取知乎回答元数据
 
     Args:
         answer_id: 回答 ID (数字字符串)
         headless: 是否使用无头浏览器
+        cookie: 知乎 Cookie（可选）
 
     Returns:
         回答元数据字典
@@ -67,25 +70,26 @@ def extract_answer(answer_id: str, headless: bool = True) -> Dict[str, Any]:
     # 回答需要构建包含问题ID的URL，但answer_id本身足以识别
     # 使用一个占位符问题ID，因为提取器会从URL中提取answer_id
     url = f"https://www.zhihu.com/question/0/answer/{answer_id}"
-    return _extract_metadata(url, "answer", headless)
+    return _extract_metadata(url, "answer", headless, cookie)
 
 
-def extract_question(question_id: str, headless: bool = True) -> Dict[str, Any]:
+def extract_question(question_id: str, headless: bool = True, cookie: str = None) -> Dict[str, Any]:
     """
     获取知乎问题元数据
 
     Args:
         question_id: 问题 ID (数字字符串)
         headless: 是否使用无头浏览器
+        cookie: 知乎 Cookie（可选）
 
     Returns:
         问题元数据字典
     """
     url = f"https://www.zhihu.com/question/{question_id}"
-    return _extract_metadata(url, "question", headless)
+    return _extract_metadata(url, "question", headless, cookie)
 
 
-def _extract_metadata(url: str, content_type: str, headless: bool) -> Dict[str, Any]:
+def _extract_metadata(url: str, content_type: str, headless: bool, cookie: str = None) -> Dict[str, Any]:
     """
     内部函数：提取元数据
 
@@ -93,11 +97,12 @@ def _extract_metadata(url: str, content_type: str, headless: bool) -> Dict[str,
         url: 知乎 URL
         content_type: 内容类型 (zvideo/zhuanlan)
         headless: 是否使用无头浏览器
+        cookie: 知乎 Cookie（可选）
 
     Returns:
         元数据字典
     """
-    extractor = ZhihuPlaywrightExtractor(headless=headless)
+    extractor = ZhihuPlaywrightExtractor(headless=headless, cookie=cookie)
     result = extractor.extract(url)
 
     if result:
@@ -170,15 +175,18 @@ def main():
 
     args = parser.parse_args()
 
-    # 提取元数据
+    # 获取知乎 Cookie（优先使用环境变量，其次使用参数）
+    cookie = os.environ.get('ZHIHU_COOKIE')
+    
+    # 提取元数据（传入 Cookie）
     if args.type == "zvideo":
-        result = extract_zvideo(args.id, headless=args.headless)
+        result = extract_zvideo(args.id, headless=args.headless, cookie=cookie)
     elif args.type == "zhuanlan":
-        result = extract_zhuanlan(args.id, headless=args.headless)
+        result = extract_zhuanlan(args.id, headless=args.headless, cookie=cookie)
     elif args.type == "answer":
-        result = extract_answer(args.id, headless=args.headless)
+        result = extract_answer(args.id, headless=args.headless, cookie=cookie)
     elif args.type == "question":
-        result = extract_question(args.id, headless=args.headless)
+        result = extract_question(args.id, headless=args.headless, cookie=cookie)
     else:
         print(f"不支持的类型: {args.type}")
         sys.exit(1)