Ultra MCP

Instructions

Implementation Reference

• src/handlers/ai-tools.ts:265-296 (handler)

  The core handler method in AIToolHandlers class that implements the 'list-ai-models' tool logic. It retrieves available models and configured providers from ProviderManager, formats a markdown list with status indicators, and returns the response in MCP format.

  async handleListModels() {
    const availableModels = this.providerManager.getAvailableModels();
    const configuredProviders = await this.providerManager.getConfiguredProviders();
  
    let response = "## Available AI Models\n\n";
    
    for (const [provider, models] of Object.entries(availableModels)) {
      const isConfigured = configuredProviders.includes(provider);
      response += `### ${provider.charAt(0).toUpperCase() + provider.slice(1)} ${isConfigured ? "✅" : "❌"}\n`;
      
      if (models.length > 0) {
        response += models.map(model => `- ${model}`).join("\n");
      } else {
        response += "- Not configured";
      }
      response += "\n\n";
    }
  
    response += `\n## Default Models\n`;
    response += `- OpenAI/Azure: gpt-5 (optimized for reasoning)\n`;
    response += `- Gemini: gemini-2.5-pro (with Google Search enabled)\n`;
    response += `- Grok: grok-4 (latest xAI model with reasoning support)\n`;
  
    return {
      content: [
        {
          type: "text",
          text: response,
        },
      ],
    };
  }

• src/server.ts:274-282 (registration)

  Registers the 'list-ai-models' tool with the MCP server using server.registerTool, providing title, description, empty input schema, and an inline handler that delegates to AIToolHandlers.handleListModels() via getHandlers().

  // Register list-ai-models tool
  server.registerTool("list-ai-models", {
    title: "List AI Models",
    description: "List all available AI models and their configuration status",
    inputSchema: {},
  }, async () => {
    const aiHandlers = await getHandlers();
    return await aiHandlers.handleListModels();
  });

• src/handlers/ai-tools.ts:1487-1493 (schema)

  Defines the tool schema and metadata for 'list-ai-models' (empty input schema) as part of the getToolDefinitions() method export, likely used for MCP tool discovery.

    name: "list-ai-models",
    description: "List all available AI models and their configuration status",
    inputSchema: {
      type: "object",
      properties: {},
    },
  },

• src/server.ts:195-235 (helper)

  The getHandlers() function lazily initializes and returns the AIToolHandlers instance, loading configs and ProviderManager, which is called by the tool registration to access handleListModels().

  async function getHandlers() {
    if (!handlers) {
      const { ConfigManager } = require("./config/manager");
      const { ProviderManager } = require("./providers/manager");
      const { AIToolHandlers } = require("./handlers/ai-tools");
      
      const configManager = new ConfigManager();
      
      // Load config and set environment variables
      const config = await configManager.getConfig();
      if (config.openai?.apiKey) {
        process.env.OPENAI_API_KEY = config.openai.apiKey;
      }
      if (config.openai?.baseURL) {
        process.env.OPENAI_BASE_URL = config.openai.baseURL;
      }
      if (config.google?.apiKey) {
        process.env.GOOGLE_API_KEY = config.google.apiKey;
      }
      if (config.google?.baseURL) {
        process.env.GOOGLE_BASE_URL = config.google.baseURL;
      }
      if (config.azure?.apiKey) {
        process.env.AZURE_API_KEY = config.azure.apiKey;
      }
      if (config.azure?.baseURL) {
        process.env.AZURE_BASE_URL = config.azure.baseURL;
      }
      if (config.xai?.apiKey) {
        process.env.XAI_API_KEY = config.xai.apiKey;
      }
      if (config.xai?.baseURL) {
        process.env.XAI_BASE_URL = config.xai.baseURL;
      }
      
      providerManager = new ProviderManager(configManager);
      handlers = new AIToolHandlers(providerManager);
    }
    
    return handlers;
  }