Cursor용 모델 컨텍스트 프로토콜(MCP) 서버로 설계된 AI 기반 개발 툴킷에 오신 것을 환영합니다! 이 프로젝트는 맞춤형 AI 도구를 통해 지능형 코딩 지원을 제공합니다. 이 글은 주로 튜토리얼 데모이며, 운영 환경에 적합한 도구는 아님을 유의하세요.

✨ 특징

🎨 코드 아키텍트

고급 추론 LLM을 호출해 코딩 에이전트를 위한 계획과 지침을 생성하세요.

📸 스크린샷 친구

UI 디자인 스크린샷을 찍어 composer 에이전트와 함께 사용하세요.

🔍 코드 리뷰

git diffs를 사용해 코드 리뷰를 트리거하세요.

🚀 시작하기

1. 환경 설정

먼저, 환경 변수를 설정해야 합니다. src/env/keys.ts에 파일을 생성하세요:

⚠️ 보안 주의: API 키를 소스 코드에 직접 저장하는 것은 운영 환경에서는 권장되지 않습니다. 이는 로컬 개발 및 학습 목적에서만 사용됩니다. 커서 MCP 인터페이스에서 env var를 인라인으로 설정할 수도 있습니다.

2. 설치

3. 서버 구축

4. 커서 추가

이 프로젝트는 커서에서 MCP 서버로 사용하도록 설계되었습니다. 설정 방법은 다음과 같습니다:

1. 커서 열기
2. [[T1]로 이동
3. 클릭 + Add New MCP Server
4. 양식 작성:

🤖 AI Development Assistant MCP Server

Welcome to your AI-powered development toolkit, designed as a Model Context Protocol (MCP) server for Cursor! This project provides intelligent coding assistance through custom AI tools. Note that this is mostly a tutorial demo, and not a production-ready tool.

✨ Features

🎨 Code Architect

Call advanced reasoning LLMs to generate plans and instructions for coding agents.

📸 Screenshot Buddy

Take UI design screenshots and use them with the composer agent.

🔍 Code Review

Use git diffs to trigger code reviews.

🚀 Getting Started

1. Environment Setup

First, you’ll need to set up your environment variables. Create a file at src/env/keys.ts:

export const OPENAI_API_KEY = "your_key_here";
// Add any other keys you need

⚠️ Security Note: Storing API keys directly in source code is not recommended for production environments. This is only for local development and learning purposes. You can set the env var inline in the Cursor MCP interface as well.

2. Installation

npm install
# or
yarn install

3. Build the Server

npm run build

4. Adding to Cursor

This project is designed to be used as an MCP server in Cursor. Here’s how to set it up:

1. Open Cursor
2. Go to Cursor Settings > Features > MCP
3. Click + Add New MCP Server
4. Fill out the form:
   • Name: AI Development Assistant
   • Type: stdio
   • Command: node /path/to/your/project/dist/index.js

📘 Pro Tip: You might need to use the full path to your project’s built index.js file.

After adding the server, you should see your tools listed under “Available Tools”. If not, try clicking the refresh button in the top right corner of the MCP server section.

For more details about MCP setup, check out the Cursor MCP Documentation.

🛠️ Using the Tools

Once configured, you can use these tools directly in Cursor’s Composer. The AI will automatically suggest using relevant tools, or you can explicitly request them by name or description.

For example, try typing in Composer:

• “Review this code for best practices”
• “Help me architect a new feature”
• “Analyze this UI screenshot”

The agent will ask for your approval before making any tool calls.

📘 Pro Tip: You can update your .cursorrules file with instructions on how to use the tools for certain scenarios, and the agent will use the tools automatically.

📁 Project Structure

src/
├── tools/
│   ├── architect.ts    # Code structure generator
│   ├── screenshot.ts   # Screenshot analysis tool
│   └── codeReview.ts   # Code review tool
├── env/
│   └── keys.ts         # Environment configuration (add your API keys here!)
└── index.ts           # Main entry point

🤝 Contributing

Contributions welcome! Please feel free to submit a Pull Request.

📝 License