@@ -0,0 +1,110 @@
+import re
+from typing import List, Dict
+
+class DailyAIConsultingNews:
+    def __init__(self):
+        self.title = ""
+        self.date = ""
+        self.body = ""
+        self.items = []
+
+    def parse(self) -> None:
+        """Parse the body content into structured news items."""
+        if not self.body:
+            self.items = []
+            return
+
+        # Split body into lines and strip whitespace
+        lines = [line.strip() for line in self.body.split('\n')]
+
+        # Extract title and date from header
+        title_line = next((line for line in lines if line.startswith('##')), '')
+        if title_line:
+            self.title = title_line[2:].strip()
+        
+        date_match = re.search(r'\d{4} 年 \d{1,2} 月 \d{1,2} 日', self.body)
+        if date_match:
+            self.date = date_match.group()
+
+        # Find the start of news items (after "## 今日 AI 咨询行业要点")
+        item_start_idx = -1
+        for i, line in enumerate(lines):
+            if line.startswith('## 今日 AI 咨询行业要点'):
+                item_start_idx = i + 1
+                break
+
+        if item_start_idx == -1:
+            self.items = []
+            return
+
+        # Collect all lines from item start onward
+        content_lines = lines[item_start_idx:]
+        items = []
+        current_item = None
+
+        reference_pattern = re.compile(r'^[-*] 参考：$|^\[参考\]$')
+
+        for line in content_lines:
+            # Match numbered headlines (### 1., ### 2., etc.)
+            headline_match = re.match(r'^###\s*\d+\.\s*(.+)$', line)
+            if headline_match:
+                if current_item is not None:
+                    items.append(current_item)
+                current_item = {
+                    'id': len(items) + 1,
+                    'headline': headline_match.group(1).strip(),
+                    'summary': '',
+                    'references': []
+                }
+                continue
+
+            if current_item is None:
+                continue
+
+            # Check for reference lines
+            if re.match(r'^[-*]\s*参考：', line) or line.startswith('Reference:'):
+                # Extract URL from markdown link if present
+                url_match = re.search(r'\[.*?\]\((https?://[^\)]+)\)', line)
+                if url_match:
+                    current_item['references'].append(url_match.group(1))
+                else:
+                    # Fallback: extract any URL
+                    url_only = re.search(r'(https?://[^\s]+)', line)
+                    if url_only:
+                        current_item['references'].append(url_only.group(1))
+                continue
+
+            # Accumulate summary text (skip separators)
+            if line in ['---', '--', '']:
+                continue
+
+            # If no current item, skip
+            if not current_item:
+                continue
+
+            # Add to summary if not a new section
+            if not line.startswith('###'):
+                if current_item['summary']:
+                    current_item['summary'] += ' ' + line
+                else:
+                    current_item['summary'] = line
+
+        # Append last item
+        if current_item is not None:
+            items.append(current_item)
+
+        self.items = items
+
+    def get_items(self) -> List[Dict]:
+        """Return parsed news items."""
+        if not self.items:
+            self.parse()
+        return self.items
+
+    def to_dict(self) -> Dict:
+        """Convert to dictionary format."""
+        return {
+            'title': self.title,
+            'date': self.date,
+            'items': self.get_items()
+        }