WebSearch

Implementation Reference

• main.py:80-86 (handler)

  Handler function for the 'scrape' tool, registered via @mcp.tool() decorator. It calls the scrape_urls helper on the WebTools instance to perform the scraping.

  @mcp.tool()
  async def scrape(url: str) -> str:
      try:
          url_scraped = webtools.scrape_urls(url)
          return url_scraped
      except Exception as e:
          return f"Error scraping url {url}: {str(e)}"

• tools/webtools.py:58-71 (helper)

  Helper method in WebTools class that implements the core scraping logic using FirecrawlApp.scrape_url, supporting markdown and HTML formats with screenshot action.

  def scrape_urls(self, url: list[str]):
      try:
          urls_scraped = self.firecrawl.scrape_url(
              url,
              params={
                  "formats": ["markdown", "html"],
                  "actions": [
                      {"type": "screenshot"},
                  ],
              },
          )
          return urls_scraped
      except Exception as e:
          return f"Error scrapping ulr {url}: {str(e)}"