mcp-searcher
A template for creating MCP servers in Node.js/TypeScript with a modular architecture. It simplifies adding new tools and includes an example Elasticsearch search tool.
README
MCP Server Template
Template hexagonal para crear MCP servers en Node.js/TypeScript. Objetivo: agregar un nuevo tool = crear 1 archivo + 1 línea.
Estructura
src/
├── index.ts ← entrypoint (3 líneas)
│
├── core/
│ ├── registry.ts ← ToolRegistry: registro + dispatcher automático
│ ├── server.ts ← factory de instancias McpServer
│ └── run.ts ← transporte dual stdio/HTTP
│
├── tools/
│ ├── index.ts ← ← ← ÚNICO ARCHIVO QUE EDITAS AL AGREGAR TOOLS
│ └── examples/
│ └── es-search.tool.ts ← ejemplo completo (patrón a seguir)
│
├── infra/
│ ├── clients/
│ │ └── elasticsearch.client.ts ← singleton de ES
│ └── formatters/
│ └── result.formatter.ts ← limpieza de resultados para el LLM
│
├── config/
│ ├── env.ts ← variables de entorno validadas con Zod
│ └── logger.ts ← pino → stderr (MCP-safe)
│
└── shared/
├── types/
│ └── tool.types.ts ← ToolDefinition interface
└── errors/
└── mcp.error.ts ← toMcpError / toMcpSuccess helpers
Agregar un nuevo tool (3 pasos)
Paso 1 — Crea src/tools/mi-feature.tool.ts
import { z } from "zod";
import { ToolDefinition } from "../shared/types/tool.types.js";
import { toMcpSuccess } from "../shared/errors/mcp.error.js";
const MiInput = z.object({
param: z.string().describe("Descripción para el LLM"),
});
async function miHandler(input: z.infer<typeof MiInput>) {
// tu lógica aquí
return toMcpSuccess({ resultado: input.param });
}
export const miTool: ToolDefinition<z.infer<typeof MiInput>> = {
name: "mi_tool",
description: "Descripción para el LLM",
inputSchema: MiInput,
handler: miHandler,
};
Paso 2 — Regístralo en src/tools/index.ts
import { miTool } from "./mi-feature.tool.js";
export function registerAllTools(registry: ToolRegistry): void {
registry
.register(esSearchKeywordTool)
.register(miTool); // ← esta línea
}
Paso 3 — Listo. El dispatcher, JSON Schema y listado MCP se actualizan solos.
Variables de entorno
SERVER_NAME=mi-mcp-server
SERVER_VERSION=1.0.0
MCP_TRANSPORT=stdio # stdio | http
PORT=3001
ELASTICSEARCH_URL=http://localhost:9200
LOG_LEVEL=info
Comandos
npm install
npm run dev # dev stdio
npm run dev:http # dev MCP over HTTP
npm run prod # prod MCP over HTTP
Regla de oro del MCP logging
Nunca uses console.log en un MCP server en modo stdio.
stdout es el canal del protocolo. Un console.log rompe el handshake.
Usa siempre logger.info(...) de config/logger.ts — escribe a stderr.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.