Update app.py

app.py

@@ -1,12 +1,10 @@
 from smolagents import CodeAgent, HfApiModel, tool
 from tools.final_answer import FinalAnswerTool
-from tools.visit_webpage import VisitWebpageTool
 from Gradio_UI import GradioUI
 import requests
 import yaml
 import os
 from typing import Dict, List, Optional
-import re  # Add this import at the top with other imports
 
 @tool
 def fetch_news(topic: str, num_results: int = 5) -> List[Dict]:
@@ -57,62 +55,12 @@ def fetch_news(topic: str, num_results: int = 5) -> List[Dict]:
     except Exception as e:
         return f"Error: {str(e)}"
 
-@tool
-def scrape_articles(articles: List[Dict]) -> List[Dict]:
-    """Scrapes the full content of news articles from their URLs.
-    
-    Args:
-        articles: List of article dictionaries containing article information
-    
-    Returns:
-        List of articles with additional full_content field
-    """
-    webpage_tool = VisitWebpageTool()
-    
-    for article in articles:
-        try:
-            # Skip known paywalled sites
-            domain = article['link'].lower()
-            if any(site in domain for site in ['nytimes.com', 'wsj.com', 'ft.com']):
-                article['full_content'] = f"Content not accessible - {article['source']} article requires subscription"
-                continue
-                
-            full_content = webpage_tool.forward(article['link'])
-            if full_content and len(full_content.strip()) > 0:
-                article['full_content'] = full_content
-            else:
-                article['full_content'] = article['snippet']
-        except Exception as e:
-            article['full_content'] = article['snippet']
-    
-    return articles
-
-# Load prompt templates
-with open("prompts.yaml", 'r') as stream:
-    prompt_templates = yaml.safe_load(stream)
-
-# Initialize the models
-qwen_model = HfApiModel(
-    max_tokens=2096,
-    temperature=0.5,
-    model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
-    custom_role_conversions=None,
-)
-
-deepseek_model = HfApiModel(
-    max_tokens=2096,
-    temperature=0.3,  # Lower temperature for more focused summaries
-    model_id='deepseek-ai/DeepSeek-R1-Distill-Qwen-32B',
-    custom_role_conversions=None,
-)
-
 @tool
 def summarize_news(articles: List[Dict]) -> str:
     """Creates a summary of the news articles followed by a list of sources.
     
     Args:
-        articles: List of article dictionaries containing news article information including title,
-                 source, date, link, snippet, and full_content fields
+        articles: List of article dictionaries containing title, source, date, link, and snippet
     
     Returns:
         A string containing a summary followed by article references
@@ -120,48 +68,21 @@ def summarize_news(articles: List[Dict]) -> str:
     if not articles or not isinstance(articles, list):
         return "No articles to summarize"
     
-    content_to_summarize = ""
-    for article in articles:
-        content = article.get('full_content', article['snippet'])
-        content_to_summarize += f"Title: {article['title']}\nSource: {article['source']}\nDate: {article['date']}\nLink: {article['link']}\nContent: {content}\n\n"
-    
-    summary_prompt = f"""Create a clear news summary from these articles. Focus on the main facts and developments.
-
-For each article, extract the most newsworthy information and remove any redundant content, formatting artifacts, or navigation elements.
-
-Present the information in this format:
-1. Start with a brief overview of the main story
-2. Include specific details and numbers when relevant
-3. Add source links in markdown format: [Source Name](URL)
-
-{content_to_summarize}"""
+    # Collect all snippets for the overall summary
+    all_snippets = [article['snippet'] for article in articles if article.get('snippet')]
     
-    try:
-        summary = deepseek_model.complete(summary_prompt).strip()
-        # Ensure proper markdown link formatting
-        for article in articles:
-            source = article['source']
-            link = article['link']
-            summary = summary.replace(f"[{source}]", f"[{source}]({link})")
-        return summary
-    except Exception as e:
-        print(f"DeepSeek summarization failed: {str(e)}")
-        return original_summary_format(articles)
-
-def original_summary_format(articles: List[Dict]) -> str:
-    # Original summary format as fallback
+    # Create a high-level summary from snippets
     summary = "📰 Summary:\n"
     summary += "Latest news covers " + ", ".join(set(article['source'] for article in articles)) + ". "
-    all_snippets = [article.get('full_content', article['snippet']) for article in articles]
     summary += "Key points: " + ". ".join(all_snippets[:2]) + "\n\n"
     
+    # List individual articles
     summary += "🔍 Articles:\n"
     for idx, article in enumerate(articles, 1):
         title = article['title']
         link = article['link']
         date = article['date']
-        content = article.get('full_content', article['snippet'])
-        snippet = content[:200] + "..." if len(content) > 200 else content
+        snippet = article['snippet'][:100] + "..." if len(article['snippet']) > 100 else article['snippet']
         
         summary += f"{idx}. **{title}**\n"
         summary += f"   {snippet}\n"
@@ -169,12 +90,24 @@ def original_summary_format(articles: List[Dict]) -> str:
     
     return summary
 
+# Load prompt templates
+with open("prompts.yaml", 'r') as stream:
+    prompt_templates = yaml.safe_load(stream)
+
+# Initialize the model
+model = HfApiModel(
+    max_tokens=2096,
+    temperature=0.5,
+    model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
+    custom_role_conversions=None,
+)
+
 final_answer = FinalAnswerTool()
 
 # Create the agent with all tools
 agent = CodeAgent(
-    model=qwen_model,  # Use Qwen model for main agent
-    tools=[fetch_news, scrape_articles, summarize_news, final_answer],
+    model=model,
+    tools=[fetch_news, summarize_news, final_answer],
     max_steps=6,
     verbosity_level=1,
     grammar=None,