Vibe Stack MCP
Helps developers choose optimal technology stacks through progressive questioning about project requirements, budget, and technical comfort level. Provides personalized recommendations focused on beginner-friendly Platform-as-a-Service solutions with deployment guides and cost estimates.
README
Vibe Stack MCP 🚀
A Model Context Protocol (MCP) server that helps VibeCoder developers choose the perfect tech stack for their projects through progressive questioning and personalized recommendations.
Perfect for rapid prototyping and getting tech stack recommendations based on your project requirements, budget, timeline, and technical comfort level.
What it does
This MCP server helps developers quickly choose optimal technology stacks without getting bogged down in technical analysis paralysis. Instead of researching dozens of frameworks, it asks smart questions about your project and gives you the modern, battle-tested stack that fits your needs.
Key Features
- Jargon-free questioning: Uses simple language anyone can understand
- Progressive elicitation: Gathers requirements step-by-step using MCP's elicitation spec
- Platform-as-a-Service focus: Recommends easy-to-use platforms like Vercel, Supabase, Netlify
- Practical guidance: Provides deployment guides and cost estimates
- Beginner-friendly: Focuses on tools that minimize technical complexity
Installation
- Install dependencies:
uv pip install fastmcp
- Run the server:
python run_server.py
- Add to your MCP client configuration (e.g., Claude Desktop):
{
"mcpServers": {
"vibe-stack-planner": {
"command": "python",
"args": ["/path/to/vibe_coder_stack_planner/run_server.py"]
}
}
}
Tools Available
start_project_planning()
Initiates the interactive planning process through a series of simple questions:
- What are you trying to build?
- Who will use it and how?
- What features do you need?
- How many users do you expect?
- What's your budget and technical comfort level?
recommend_stack(session_requirements?)
Provides tech stack recommendations based on gathered requirements. Can be used with or without the interactive process.
explain_recommendation(detail_level?)
Explains why specific technologies were recommended, with "basic" or "detailed" explanations.
get_deployment_guide(platform?)
Provides step-by-step deployment instructions tailored to your specific needs.
Example Usage
User: I want to build something but I don't know where to start technically.
AI: Let me help you plan the right tech stack! I'll use the vibe coder stack planner to ask you some simple questions.
[Uses start_project_planning tool]
Server: Let's start planning your project! First, tell me about your vision.
What problem are you trying to solve, or what idea do you want to build?
[Progressive questioning continues...]
Server: 🎉 Perfect! Based on what you've told me, here's my recommendation:
**Recommended Tech Stack:**
• Frontend: Next.js (React framework) for a modern web app
• Backend: Supabase (handles database, auth, and API automatically)
• Hosting: Vercel (free tier covers most small projects)
• Domain: Namecheap or Google Domains (~$12/year)
**Why this works for you:**
I chose beginner-friendly tools that handle most technical details automatically. These tools let you build and deploy quickly. This stack has generous free tiers to keep costs minimal.
Architecture
The server uses:
- FastMCP: High-level Python framework for MCP servers
- Elicitation Spec: Latest MCP elicitation specification for interactive questioning
- Rule-based recommendations: Analyzes requirements to suggest appropriate technologies
- Progressive disclosure: Builds complexity gradually based on user comfort level
Supported Platforms
The server focuses on Platform-as-a-Service solutions:
- Frontend: Vercel, Netlify, GitHub Pages
- Backend: Vercel Functions, Netlify Functions, Supabase
- Database: Supabase, PlanetScale, Firebase
- Auth: Supabase Auth, Auth0, Clerk
- Hosting: Vercel, Netlify, Render
Development
To extend or modify the server:
- Add new question types: Modify the elicitation flow in
_ask_about_*functions - Enhance recommendations: Update the
_analyze_requirementsfunction - Add new platforms: Extend the recommendation logic and deployment guides
- Improve UI: The elicitation spec supports rich form controls
License
This project is open source. Feel free to fork, extend, and contribute!
Built with ❤️ for the vibe coder community
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。