DevLens MCP

Instructions

Implementation Reference

• src/devlens/tools/advanced.py:159-220 (handler)

  The actual implementation of the link extraction logic.

  async def extract_links(url: str, *, filter_external: bool = True) -> str:
      """Extract all links from a page.
  
      Args:
          url: URL to extract links from.
          filter_external: Only return same-domain links.
  
      Returns:
          Markdown list of links organized by type.
  
      Example:
          >>> links = await extract_links("https://example.com")
      """
      try:
          import httpx
          from bs4 import BeautifulSoup
  
          async with httpx.AsyncClient(timeout=15, follow_redirects=True) as client:
              resp = await client.get(url)
              resp.raise_for_status()
              html = resp.text
  
          soup = BeautifulSoup(html, "html.parser")
          base_domain = urlparse(url).netloc
  
          # Categorize links
          from urllib.parse import urljoin
  
          internal_links: list[tuple[str, str]] = []  # (url, text)
          external_links: list[tuple[str, str]] = []
  
          for a in soup.find_all("a", href=True):
              href = a["href"]
              text = a.get_text(strip=True) or href
              absolute_url = urljoin(url, href)
              parsed = urlparse(absolute_url)
  
              if parsed.scheme in ("http", "https"):
                  if parsed.netloc == base_domain:
                      internal_links.append((absolute_url, text))
                  else:
                      external_links.append((absolute_url, text))
  
          # Build report
          report_lines = [
              f"# Links from {url}\n",
              f"## Internal Links ({len(internal_links)})\n",
          ]
  
          # Deduplicate and sort
          internal_links = sorted(set(internal_links), key=lambda x: x[1].lower())
          external_links = sorted(set(external_links), key=lambda x: x[1].lower())
  
          for link_url, text in internal_links[:50]:  # Limit to 50
              report_lines.append(f"- [{text}]({link_url})")
  
          if not filter_external and external_links:
              report_lines.append(f"\n## External Links ({len(external_links)})\n")
              for link_url, text in external_links[:30]:  # Limit to 30
                  report_lines.append(f"- [{text}]({link_url})")
  
          return "\n".join(report_lines)

• src/devlens/server.py:147-159 (registration)

  The MCP tool wrapper function that calls extract_links.

  async def tool_extract_links(url: str, filter_external: bool = True) -> str:
      """Extract all links from a page.
  
      Useful for discovering navigation structure and resources.
  
      Args:
          url: URL to extract links from.
          filter_external: Only return same-domain links (default True).
  
      Returns:
          Organized list of internal and external links.
      """
      return await extract_links(url, filter_external=filter_external)