MCP Server Template for Cursor IDE

Instructions

Implementation Reference

• mcp_simple_tool/server.py:99-131 (handler)

  Core handler function that asynchronously fetches the website using httpx, follows redirects, handles timeouts, HTTP errors, and exceptions, returning the HTML content or error as TextContent.

  async def fetch_website(
      url: str,
  ) -> list[types.TextContent | types.ImageContent | types.EmbeddedResource]:
      headers = {
          "User-Agent": "MCP Test Server (github.com/modelcontextprotocol/python-sdk)"
      }
      try:
          timeout = httpx.Timeout(10.0, connect=5.0)
          async with httpx.AsyncClient(
              follow_redirects=True, 
              headers=headers,
              timeout=timeout
          ) as client:
              response = await client.get(url)
              response.raise_for_status()
              return [types.TextContent(type="text", text=response.text)]
      except httpx.TimeoutException:
          return [types.TextContent(
              type="text",
              text="Error: Request timed out while trying to fetch the website."
          )]
      except httpx.HTTPStatusError as e:
          return [types.TextContent(
              type="text",
              text=(f"Error: HTTP {e.response.status_code} "
                    "error while fetching the website.")
          )]
      except Exception as e:
          return [types.TextContent(
              type="text",
              text=f"Error: Failed to fetch website: {str(e)}"
          )]

• mcp_simple_tool/server.py:208-214 (handler)

  Dispatch logic in the call_tool handler that routes 'mcp_fetch' calls to the fetch_website function after validating the 'url' argument.

  if name == "mcp_fetch":
      if "url" not in arguments:
          return [types.TextContent(
              type="text",
              text="Error: Missing required argument 'url'"
          )]
      return await fetch_website(arguments["url"])

• mcp_simple_tool/server.py:252-260 (schema)

  Input schema defining the required 'url' parameter as a string for the mcp_fetch tool.

      "type": "object",
      "required": ["url"],
      "properties": {
          "url": {
              "type": "string",
              "description": "URL to fetch",
          }
      },
  },

• mcp_simple_tool/server.py:248-261 (registration)

  Tool registration in list_tools() that defines the name, description, and schema for 'mcp_fetch'.

  types.Tool(
      name="mcp_fetch",
      description="Fetches a website and returns its content",
      inputSchema={
          "type": "object",
          "required": ["url"],
          "properties": {
              "url": {
                  "type": "string",
                  "description": "URL to fetch",
              }
          },
      },
  ),