MCP2Tavily

Instructions

Args:
    url (str): The URL to extract content from
    
Returns:
    str: The extracted content from the URL

Implementation Reference

• mcp2tavily.py:111-121 (handler)

  Handler function for the 'get_url_content' tool, registered via @mcp.tool() decorator. It defines the input schema via type hints and docstring, and delegates to the internal _get_url_content implementation.

  @mcp.tool()
  def get_url_content(url: str) -> str:
      """Get the content from a specific URL using Tavily API
      
      Args:
          url (str): The URL to extract content from
          
      Returns:
          str: The extracted content from the URL
      """
      return _get_url_content(url)

• mcp2tavily.py:62-99 (helper)

  The main helper function containing the tool logic: uses TavilyClient.extract() to fetch URL content, processes it with UTF-8 handling, adds metadata, and includes error handling.

  def _get_url_content(url: str) -> str:
      """Internal function to get content from a specific URL using Tavily API"""
      try:
          tavily_client = TavilyClient(api_key=API_KEY)
          logger.info(f"Attempting to extract content from URL: {url}")
          
          response = tavily_client.extract(url)
          # logger.info(f"Raw API response: {response}")  # 使用 logger 替代 print
          
          # 处理返回的数据结构
          results = response.get('results', [])
          if not results:
              logger.error(f"No results found in response: {response}")
              return "No content found for this URL. API response contains no results."
              
          # 获取第一个结果的原始内容
          first_result = results[0]
          # logger.info(f"First result structure: {list(first_result.keys())}")  # 只记录键名，避免日志过大
          
          content = first_result.get('raw_content', '')
          if not content:
              logger.error("No raw_content found in first result")
              return "No raw content available in the API response"
          
          # 确保响应文本是UTF-8编码
          content = content.encode('utf-8').decode('utf-8')
          
          # 添加一些元数据到输出中
          metadata = f"URL: {url}\n"
          metadata += f"Content length: {len(content)} characters\n"
          metadata += "---\n\n"
          
          logger.info(f"Successfully extracted content with length: {len(content)}")
          return f"{metadata}{content}"
          
      except Exception as e:
          logger.exception(f"Detailed error while extracting URL content")
          return f"Error getting content from URL: {str(e)}"