Playwright MCP

Instructions

Implementation Reference

• src/playwright_mcp/server.py:265-268 (handler)

  Handler for the 'get_page_content' tool: retrieves the active page and returns its HTML content using Playwright's page.content() method.

  elif name == "get_page_content":
      page = get_active_page(arguments.get("page_id"))
      content = await page.content()
      return [types.TextContent(type="text", text=content)]

• src/playwright_mcp/server.py:134-143 (registration)

  Registers the 'get_page_content' tool in the list_tools() function, including its name, description, and input schema.

  types.Tool(
      name="get_page_content",
      description="Get the current page HTML content",
      inputSchema={
          "type": "object",
          "properties": {
              "page_id": {"type": "string"},
          },
      },
  ),

• src/playwright_mcp/server.py:134-143 (schema)

  Defines the input schema for 'get_page_content' tool: optional page_id string.

  types.Tool(
      name="get_page_content",
      description="Get the current page HTML content",
      inputSchema={
          "type": "object",
          "properties": {
              "page_id": {"type": "string"},
          },
      },
  ),

• src/playwright_mcp/server.py:200-211 (helper)

  Helper function used by the handler to retrieve the page instance based on page_id or current default.

  def get_active_page(page_id: Optional[str] = None) -> Page:
      """Get the active page based on page_id or current default."""
      global current_page_id
      
      if page_id is None:
          page_id = current_page_id
      
      if page_id not in pages:
          raise ValueError(f"Page not found: {page_id}")
      
      return pages[page_id]