mcpField
N
NestJS MCP Server Module

NestJS MCP Server Module

Local
github.com/rekog-labsDeveloper Tools725Updated today

About NestJS MCP Server Module

NestJS MCP Server Module (@rekog/mcp-nest) is a NestJS module that exposes methods from a NestJS application as MCP tools, resources and prompts. It runs as a NestJS microservice transport strategy, so MCP handlers use the framework's guards, pipes and filters. It suits teams building enterprise MCP servers on NestJS.

Key features

  1. 01Runs as a NestJS CustomTransportStrategy
  2. 02Streamable HTTP and STDIO transports
  3. 03Tool input validation with Zod
  4. 04Guard-based authentication and per-tool authorization
  5. 05Resource templates and prompts

Use cases

  1. 01Exposing existing NestJS services as MCP tools
  2. 02Securing an MCP server with OAuth or an external authorization server
  3. 03Organizing MCP components with NestJS dependency injection

MCP support

This is a module for building MCP servers that provides @Tool methods, resources and prompts to MCP clients over Streamable HTTP or STDIO.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

NestJS MCP Server Module

README image

CI Code Coverage NPM Version NPM Downloads NPM License

A NestJS module to effortlessly expose tools, resources, and prompts for AI, from your NestJS applications using the Model Context Protocol (MCP).

With @rekog/mcp-nest you define tools, resources, and prompts in a way that's familiar in NestJS and leverage the full power of dependency injection to utilize your existing codebase in building complex enterprise ready MCP servers.

Features

Are you interested to build ChatGPT widgets (with the OpenAI SDK) or MCP apps? Find out how to do that with @rekog/MCP-Nest in this repository MCP-Nest-Samples

Tip

You can easily learn about this package using the chat tab in Context7. Better yet, connect the Context7 MCP server to allow your AI agents to access the documentation and implement MCP-Nest for you.

Installation

npm install @rekog/mcp-nest \
  @modelcontextprotocol/server \
  @modelcontextprotocol/core \
  @modelcontextprotocol/node \
  zod@^4
Optional dependencies

The built-in authorization server now lives in a separate package. If you use it, install it alongside @rekog/mcp-nest:

npm install @rekog/mcp-nest-auth

If you additionally use the TypeORM store for the authorization server, install the following optional peer dependencies as well:

npm install @nestjs/typeorm typeorm

Quick Start

MCP-Nest runs as a NestJS microservice transport strategy. Tools, resources, and prompts live on @McpController() classes (so NestJS guards, pipes, interceptors, and exception filters apply to them), and the strategy serves them over one or more transports (Streamable HTTP, STDIO).

// greeting.controller.ts
import { McpController, Tool, McpContext } from '@rekog/mcp-nest';
import { Ctx, Payload } from '@nestjs/microservices';
import { z } from 'zod';

@McpController()
export class GreetingController {
  @Tool({
    name: 'greeting-tool',
    description: 'Returns a greeting with progress updates',
    parameters: z.object({ name: z.string().default('World') }),
  })
  async sayHello(
    @Payload() { name }: { name: string },
    @Ctx() ctx: McpContext,
  ) {
    await ctx.reportProgress({ progress: 50, total: 100 });
    return { content: [{ type: 'text', text: `Hello, ${name}!` }] };
  }
}
// app.module.ts
import { Module } from '@nestjs/common';
import {
  McpStrategy,
  MCP_STRATEGY,
  StreamableHttpTransport,
} from '@rekog/mcp-nest';
import { GreetingController } from './greeting.controller';

// The strategy is the whole configuration — there is no McpModule.
export const mcp = new McpStrategy({
  name: 'my-mcp-server',
  version: '1.0.0',
  transports: [
    new StreamableHttpTransport(),
  ],
});

@Module({
  controllers: [GreetingController],
  // Optional: only needed if a provider injects the strategy (e.g. for
  // runtime/dynamic tool registration).
  providers: [{ provide: MCP_STRATEGY, useValue: mcp }],
})
export class AppModule {}
// main.ts
import { NestFactory } from '@nestjs/core';
import { AppModule, mcp } from './app.module';

async function bootstrap() {
  const app = await NestFactory.create(AppModule);
  mcp.setHttpAdapter(app.getHttpAdapter()); // needed for HTTP transports
  app.connectMicroservice({ strategy: mcp });
  await app.startAllMicroservices(); // mounts the MCP transports
  await app.listen(3000); // also serves your normal HTTP routes
}
void bootstrap();

Order matters: call startAllMicroservices() before listen() so the MCP HTTP routes are mounted before the server starts accepting connections.

That /mcp endpoint is dual-era: it answers both the 2025-era protocol (initialize handshake + sessions) and the stateless 2026-07-28 revision, concurrently, from the same @Tool methods. ctx.reportProgress(...) above reaches a 2026-07-28 client as-is — every modern request can stream progress back even though it is sessionless. A 2025-era client needs a session-aware transport for it (new StreamableHttpTransport({ statefulMode: true }) or StdioTransport); on the legacy stateless default it is a no-op. See Protocol Revisions.

For an STDIO-only server, skip the HTTP adapter and use NestFactory.createMicroservice(AppModule, { strategy: mcp }) with transports: [new StdioTransport()] and disable logging (stdout is reserved for the protocol).

Documentation

Examples

The examples directory contains working examples for all features. Refer to examples/README.md for details.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.