MCP Complete Implementation Guide

Instructions

Implementation Reference

• server.js:313-341 (handler)

  Handler function that fetches URL content using node-fetch and returns structured response including status, headers, and content.

  async fetchUrl(url, method = 'GET', headers = {}) {
    const fetch = require('node-fetch');
    
    try {
      const response = await fetch(url, {
        method,
        headers,
      });
  
      const content = await response.text();
      const responseInfo = {
        status: response.status,
        statusText: response.statusText,
        headers: Object.fromEntries(response.headers),
        content,
      };
  
      return {
        content: [
          {
            type: 'text',
            text: JSON.stringify(responseInfo, null, 2),
          },
        ],
      };
    } catch (error) {
      throw new Error(`Failed to fetch URL: ${error.message}`);
    }
  }

• server.py:268-290 (handler)

  Handler method that uses the requests library to perform HTTP requests and returns a JSON-structured response with status, headers, and content.

  async def fetch_url(
      self, url: str, method: str = "GET", headers: Optional[dict] = None
  ) -> list[types.TextContent]:
      """Fetch content from URL"""
      try:
          response = requests.request(
              method=method,
              url=url,
              headers=headers or {},
              timeout=30,
          )
          
          response_info = {
              "status": response.status_code,
              "statusText": response.reason,
              "headers": dict(response.headers),
              "content": response.text,
          }
          
          return [types.TextContent(type="text", text=json.dumps(response_info, indent=2))]
      except Exception as error:
          raise Exception(f"Failed to fetch URL: {str(error)}")

• server.js:116-138 (schema)

  Input schema definition for the fetch_url tool, specifying parameters url (required), method, and headers.

  {
    name: 'fetch_url',
    description: 'Fetch content from a URL',
    inputSchema: {
      type: 'object',
      properties: {
        url: {
          type: 'string',
          description: 'The URL to fetch',
        },
        method: {
          type: 'string',
          description: 'HTTP method (GET, POST, etc.)',
          default: 'GET',
        },
        headers: {
          type: 'object',
          description: 'HTTP headers to include',
        },
      },
      required: ['url'],
    },
  },

• server.py:145-166 (schema)

  Input schema definition for the fetch_url tool in the tool registration, matching the JS version.

      name="fetch_url",
      description="Fetch content from a URL",
      inputSchema={
          "type": "object",
          "properties": {
              "url": {
                  "type": "string",
                  "description": "The URL to fetch",
              },
              "method": {
                  "type": "string",
                  "description": "HTTP method (GET, POST, etc.)",
                  "default": "GET",
              },
              "headers": {
                  "type": "object",
                  "description": "HTTP headers to include",
              },
          },
          "required": ["url"],
      },
  ),

• chatgpt-proxy.js:256-269 (handler)

  Inline handler for fetch_url in the ChatGPT proxy server, similar to server.js implementation.

  case 'fetch_url':
    const response = await fetch(args.url, {
      method: args.method || 'GET',
      headers: args.headers || {},
    });
    const responseContent = await response.text();
    return {
      success: true,
      status: response.status,
      statusText: response.statusText,
      headers: Object.fromEntries(response.headers),
      content: responseContent,
    };