NestJS DevTools MCP
Give your AI coding agents the superpower to observe your NestJS application's runtime state in real-time.
README
NestJS DevTools MCP
Give your AI coding agents (Claude, Cursor, Copilot) the superpower to observe your NestJS application's runtime state in real-time.
Ever wished AI could see your crashed application logs, registered routes, or current DI container without you copy-pasting terminal outputs? This project provides a transparent, near-zero config bridge between your running NestJS app and your AI tools via the Model Context Protocol (MCP).
Current MCP Tools
Currently available tools:
discover_servers— Find local NestJS servers with the plugin enabled.get_logs— Retrieve recent runtime logs from a detected NestJS server.get_routes— List registered HTTP routes (method, path, controller, handler).get_request_history— Retrieve recent HTTP request history with filters for method, status, path, duration, and errors.get_config— Retrieve sanitized runtime configuration from environment variables and ConfigService.
get_request_history captures real HTTP traffic, including unmatched 404s, without recording request or response bodies by default. Internal /_dev/mcp/* calls are excluded so tool calls do not pollute the history.
get_config is read-only and always masks values that look sensitive, including tokens, passwords, auth keys, private keys, and database URLs. Secret masking cannot be disabled.
More tools may be added in future releases.
Quick Start (2 Steps)
Step 1: Integrate Plugin into NestJS
Install the plugin package in your NestJS application:
# Using npm
npm install @nestjs-devtools-mcp/plugin
# Using yarn
yarn add @nestjs-devtools-mcp/plugin
# Using pnpm
pnpm add @nestjs-devtools-mcp/plugin
Configure app.module.ts:
import { Module } from '@nestjs/common'
import { DevtoolsMcpModule } from '@nestjs-devtools-mcp/plugin'
@Module({
imports: [
DevtoolsMcpModule.register(), // Automatically disables in production
],
})
export class AppModule {}
Request body capture is off by default. If you need it for local debugging, enable DevtoolsMcpModule.register({ captureRequestBody: true }); multipart bodies are never captured, and configuration secrets returned by get_config are always masked.
Apply the custom logger in main.ts to allow context interception:
import { NestFactory } from '@nestjs/core'
import { AppModule } from './app.module'
import { applyDevtoolsLogger } from '@nestjs-devtools-mcp/plugin'
async function bootstrap() {
const app = await NestFactory.create(AppModule, {
bufferLogs: true,
})
// Activate DevTools logger
applyDevtoolsLogger(app)
await app.listen(3000)
}
bootstrap()
Step 2: Configure MCP Client
Add the following to your AI Assistant's MCP settings (e.g., claude_desktop_config.json):
{
"mcpServers": {
"nestjs-devtools": {
"command": "npx",
"args": ["-y", "nestjs-devtools-mcp@latest"]
}
}
}
That's it! Restart your MCP client and ask your AI: "Fetch the latest logs from my NestJS application."
Documentation & Guides
Looking for advanced setups or want to contribute? Dive deeper into our docs:
- Installation & Usage Guide - Advanced setups, installing from source, and troubleshooting.
- Development Guide - For contributors, detailing the monorepo setup, building, testing, and linting.
- API Specs - The MCP protocol specification and contract.
How it Works (Architecture)
The project uses a secure 2-package model to avoid interfering with your app's logic:
AI Client (Claude, Cursor, ...)
│
│ [STDIO - MCP Protocol]
▼
nestjs-devtools-mcp (Bridge Server via npx)
│
│ [HTTP - Localhost Only]
▼
@nestjs-devtools-mcp/plugin (Runs inside your App)
│
▼
NestJS Runtime (Logger, Container, Routes)
- The Plugin (
@nestjs-devtools-mcp/plugin) runs inside your NestJS process, safely collecting runtime data into circular buffers and exposing an internal HTTP endpoint. - The Server (
nestjs-devtools-mcp) is a lightweight Bridge CLI that manages STDIO communication with the AI and proxies tool calls over HTTP to the plugin.
License
MIT © HaoNgo232. For production environments, always reassess security assumptions before deploying plugins that observe application state.
Marketplace Ownership Verification
To improve trust signals in LobeHub MCP Marketplace and verify ownership:
- Keep the MCP badge in this README (already added above).
- Open your listing page: https://lobehub.com/mcp/hao%20ngo232-nestjs-devtools-mcp.
- Use the "Check Claim Status" flow and complete GitHub ownership verification.
After LobeHub re-crawls the repository, the owner claim status should be updated on the score page.
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.