fetch_page_markdown

Instructions

Implementation Reference

• server.py:205-269 (handler)

  The main handler function for the fetch_page_markdown tool. Decorated with @mcp.tool() to register as an MCP tool. Takes a Confluence page ID, fetches the page content via the ConfluenceClient, extracts metadata (title, space, version, URL, labels), converts the HTML body to Markdown using markdownify, and returns a formatted Markdown string with a metadata header.

  @mcp.tool()
  def fetch_page_markdown(page_id: str) -> str:
      """
      Fetch a Confluence page and convert it to Markdown format.
      
      This tool retrieves page content and converts it from HTML to Markdown,
      reducing token usage by 60-80% compared to raw HTML while preserving
      formatting, links, and structure.
      
      Args:
          page_id: The Confluence page ID to fetch
      
      Returns:
          Markdown-formatted page content with metadata header
      """
      try:
          if not page_id or not page_id.strip():
              return "Error: page_id parameter is required and cannot be empty"
          
          # Fetch page content
          page = confluence.get_page_content(page_id)
          
          # Extract metadata
          title = page.get("title", "Untitled")
          space_key = page.get("space", {}).get("key", "")
          space_name = page.get("space", {}).get("name", "")
          version = page.get("version", {}).get("number", 1)
          page_url = f"{CONFLUENCE_URL}/wiki{page.get('_links', {}).get('webui', '')}"
          
          # Extract HTML content
          html_content = page.get("body", {}).get("storage", {}).get("value", "")
          
          if not html_content:
              return f"Error: No content found for page ID {page_id}"
          
          # Convert HTML to Markdown
          markdown_content = md(
              html_content,
              heading_style="ATX",
              bullets="-",
              strip=['script', 'style']
          )
          
          # Extract labels if available
          labels = page.get("metadata", {}).get("labels", {}).get("results", [])
          label_names = [label.get("name") for label in labels if label.get("name")]
          
          # Format final output with metadata
          output = f"""# {title}
  
  **Space:** {space_name} ({space_key})  
  **Version:** {version}  
  **URL:** {page_url}
  """
          
          if label_names:
              output += f"**Labels:** {', '.join(label_names)}\n"
          
          output += f"\n---\n\n{markdown_content}"
          
          return output
      
      except Exception as e:
          logger.error(f"Error fetching page {page_id}: {e}")
          return f"Error: {str(e)}"

• server.py:205-205 (registration)

  Registration of fetch_page_markdown as an MCP tool via the @mcp.tool() decorator from the FastMCP framework.

  @mcp.tool()

• server.py:102-107 (helper)

  Helper method on ConfluenceClient that fetches page content from the Confluence API, expanding body.storage, space, version, and metadata.labels - used by fetch_page_markdown to retrieve page data.

  def get_page_content(self, page_id: str) -> dict:
      """Fetch a specific page with its content."""
      params = {
          "expand": "body.storage,space,version,metadata.labels"
      }
      return self._make_request("GET", f"/content/{page_id}", params=params)

• server.py:214-218 (schema)

  Schema/docstring defining the input parameter (page_id: str) and the return type (str) for the fetch_page_markdown tool.

  Args:
      page_id: The Confluence page ID to fetch
  
  Returns:
      Markdown-formatted page content with metadata header