Supabase MCP Server
Enables notes management through MCP integration with a Supabase PostgreSQL database, supporting full CRUD operations. It features secure user data isolation using JWT authentication and Row Level Security policies.
README
Supabase MCP Server - Notes Application
Description
A full-featured notes management application built with Model Context Protocol (MCP) integration and Supabase as the database backend. The application ensures secure user access with role-based isolation through combined protection at the application level (JWT) and database level (RLS policies).
Features
🔐 Security
- OAuth 2.0 Authentication - supports Email/Password and GitHub login
- JWT Token Validation - protects MCP endpoints
- Row Level Security (RLS) - database policies for user data isolation
- XSS Protection - HTML escape functions in client code
📝 Functionality
- CRUD Operations - Create, read, update, and delete notes
- User Association - each note is linked to its creator
- Responsive Interface - optimized for all devices
- User Management - login, registration, and logout
🏗️ Architecture
- Backend: Node.js Express server with MCP integration
- Frontend: HTML/CSS/JavaScript without external dependencies
- Database: Supabase PostgreSQL
- API: RESTful MCP endpoints with JWT authentication
Getting Started
Prerequisites
- Node.js (v14+)
- Supabase project (https://supabase.com)
- Postman for API testing (optional)
Installation
- Navigate to project directory:
cd supabase-mcp-server
- Install dependencies:
npm install
- Configure .env file:
cp .env.example .env
Fill in the following values from your Supabase project:
SUPABASE_URL=https://your-project.supabase.co
SUPABASE_ANON_KEY=your-anon-key
PORT=3000
- Start the server:
npm start
The server will start at http://localhost:3000
Usage
Frontend Application
- Open
http://localhost:3000in your browser - Choose a login option:
- Email/Password: Enter your email and password
- GitHub: Click the GitHub button
- After login, you can:
- Create notes - fill in title and content, click "Add"
- Edit notes - click "Edit", make changes, click "Save"
- Delete notes - click "Delete" and confirm
MCP Endpoints (Postman)
The application provides MCP endpoints for CRUD operations:
Postman Collection: Open postman_collection.json in Postman
Testing Steps:
- Log in to the frontend
- Copy the JWT token (displayed in the application)
- In Postman, go to Collection Variables
- Paste the token in the
{{token}}variable - Execute the requests
Technical Details
notes Table:
CREATE TABLE notes (
id BIGINT PRIMARY KEY GENERATED BY DEFAULT AS IDENTITY,
user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE,
title TEXT NOT NULL,
content TEXT,
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);
RLS Policies
Four RLS policies protect data access:
| Policy | Operation | Condition |
|---|---|---|
Users see own notes |
SELECT | auth.uid() = user_id |
Users insert own notes |
INSERT | auth.uid() = user_id |
Users update own notes |
UPDATE | auth.uid() = user_id |
Users delete own notes |
DELETE | auth.uid() = user_id |
Local Testing
- Open two browser windows - one for each user
- Log in with different accounts
- Create notes in each account
- Verify that each user sees only their own notes
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。