MySQL MCP Server

MySQL MCP Server

Enables Cursor IDE to interact with MySQL databases through schema and data operations, including read and write capabilities with environment-specific safety controls.

Category
访问服务器

README

MySQL MCP Server

A comprehensive MySQL Model Context Protocol (MCP) server that provides database operations for Cursor IDE.

Features

  • Connection Management: Connect to any MySQL database
  • Schema Operations: List databases, tables, describe table structures
  • Data Operations: SELECT, INSERT, UPDATE, DELETE operations
  • Advanced Features: View indexes, foreign keys, execute custom queries
  • Safety: Built-in query limits and error handling
  • 🔒 Production Security: Automatic PII protection and audit logging (production environment only)

Prerequisites

Before setting up this MySQL MCP server, ensure you have:

  1. Node.js (v16 or higher) - Required to run the MCP server
  2. MySQL Database - Either local or remote MySQL server access
  3. Cursor IDE - For MCP integration

📁 Repository Note: This repository follows Node.js best practices by excluding node_modules/ and dist/ folders. You'll generate these locally after cloning.

Quick Setup

🚀 First Time Setup (After Cloning)

Important: This repository does not include node_modules/ or dist/ folders to keep it lightweight. You'll need to generate them:

  1. Clone the repository:

    git clone https://github.com/binoy154/mysql-mcp.git
    cd mysql-mcp
    

    ⚠️ Important Path Considerations:

    • Note the absolute path where you clone this repository (e.g., C:/Projects/mysql-mcp or E:/mysql-mcp)
    • You'll need this exact path for Cursor MCP configuration later
    • The built files will be at YOUR_PATH/dist/index.js
  2. Install dependencies:

    npm install
    

    This will create the node_modules/ folder with all required packages.

  3. Build the project:

    npm run build
    

    This will create the dist/ folder with compiled JavaScript files.

  4. Set up environment variables (optional):

    cp .env.example .env
    # Edit .env with your MySQL credentials
    

    Note: You can also configure credentials directly in Cursor MCP settings.

  5. Configure Cursor MCP:

    🔧 IMPORTANT: Update the file path below with your actual clone location!

    A single MCP server instance can switch between local, staging, preproduction and production environments at runtime. The local environment allows full read/write access, while staging, preproduction, and production are configured as read-only for safety. Supply their credentials through environment variables:

    {
      "mcpServers": {
        "mysql": {
          "command": "node",
          "args": ["YOUR_ABSOLUTE_PATH/mysql-mcp/dist/index.js"],
          "env": {
            "MYSQL_LOCAL_HOST": "localhost",
            "MYSQL_LOCAL_PORT": "3306",
            "MYSQL_LOCAL_USER": "root",
            "MYSQL_LOCAL_PASSWORD": "localPassword",
            "MYSQL_LOCAL_DATABASE": "localDatabase",
    
            "MYSQL_STAGING_HOST": "staging.db.example.com",
            "MYSQL_STAGING_PORT": "3306",
            "MYSQL_STAGING_USER": "staging_user",
            "MYSQL_STAGING_PASSWORD": "staging_pw",
            "MYSQL_STAGING_DATABASE": "staging_db",
    
            "MYSQL_PREPRODUCTION_HOST": "preprod-db.example.com",
            "MYSQL_PREPRODUCTION_PORT": "3306",
            "MYSQL_PREPRODUCTION_USER": "preprod_user",
            "MYSQL_PREPRODUCTION_PASSWORD": "preprod_pw",
            "MYSQL_PREPRODUCTION_DATABASE": "preprod_db"
          }
        }
      }
    }
    

    📝 Path Examples - Replace YOUR_ABSOLUTE_PATH/mysql-mcp with:

    • Windows: "C:/Projects/mysql-mcp/dist/index.js"
    • Windows (E drive): "E:/mysql-mcp/dist/index.js"
    • Mac/Linux: "/home/username/mysql-mcp/dist/index.js"

    ⚠️ Critical: Use forward slashes (/) even on Windows, and ensure the path points to where you actually cloned the repository.

    Database Safety: The production credentials should point to a read-only slave or replica to guarantee safety. Additionally, staging and preproduction environments are enforced as read-only at the code level to prevent accidental data modifications.

Available Tools

Connection Tools

  • connect_database: Connect with custom credentials
  • list_databases: Show all available databases
  • use_database: Switch to a specific database
  • switch_environment: Switch the active DB environment (local, staging, preproduction, production)

Schema Tools

  • list_tables: List all tables in current database
  • describe_table: Get detailed table schema
  • get_table_indexes: View table indexes
  • get_foreign_keys: View foreign key relationships

Data Operations

  • select_query: Execute SELECT queries with automatic LIMIT
  • insert_data: Insert new records
  • update_data: Update existing records
  • delete_data: Delete records
  • execute_query: Execute any SQL query (use carefully)

Write operations & permission levels

Tool Local Staging Pre-production Production
insert_data, update_data, delete_data, execute_query ✅ Immediate READ-ONLY READ-ONLY READ-ONLY

Note: Only the local environment allows write operations. All other environments (staging, preproduction, production) are configured as read-only for data safety and integrity.

Example (local only):

{
  "name": "update_data",
  "arguments": {
    "table": "users",
    "data": { "status": "active" },
    "where": "id = 42"
  }
}

Usage Examples

Connect to Database

Use the connect_database tool with:
- host: localhost
- port: 3306
- user: myuser
- password: mypassword
- database: myapp_db

Query Data

Use select_query with query: "SELECT * FROM users WHERE active = 1"

Insert Data

Use insert_data with:
- table: "users"
- data: {"name": "John Doe", "email": "john@example.com"}

Security Notes

  1. Store database credentials securely using environment variables
  2. SELECT queries automatically include LIMIT 100 unless specified
  3. The server uses parameterized queries for data operations
  4. Ensure your MySQL user has appropriate permissions

🔒 Production Security Features

Automatic PII Protection (Production Environment Only):

  • Sensitive Data Masking: Columns containing PII (SIN, birth dates, credit cards, etc.) are automatically masked
  • Audit Logging: All production database access is logged for compliance
  • Zero Impact on Development: Local, staging, and preproduction environments work exactly as before
  • Smart Detection: Uses pattern matching to identify sensitive columns (sin, ssn, birth_date, etc.)

Environment Behavior:

  • Local/Staging/Preproduction: Full data access (no filtering)
  • Production: Automatic sensitive data masking + audit logging

Example Output in Production:

{
  "id": 1,
  "name": "John Doe", 
  "email": "jo***@example.com",
  "sin": "12*8",
  "birth_date": "19**-**-**"
}

Security Status Messages:

  • 🔓 DEVELOPMENT MODE: No data filtering (full access)
  • 🔒 PRODUCTION MODE: Sensitive data protection active

Deployment to Another Computer

❌ What Won't Work

Simply copying the project folder to another computer will not work without proper setup.

✅ Required Steps for New Installation

  1. Install Prerequisites:

    • Node.js (v16 or higher)
    • Access to a MySQL database
  2. Clone or Copy Project Files:

    git clone https://github.com/binoy154/mysql-mcp.git
    cd mysql-mcp
    
    • Note the absolute path where you cloned/copied the project
    • You'll need this exact path for Cursor MCP configuration
  3. Install Dependencies:

    npm install
    
  4. Build the Project:

    npm run build
    
  5. Update Cursor Configuration:

    • Update the file path in your Cursor MCP settings to match the new location
    • Example: Change "E:/mySqlMcp/dist/index.js" to the correct path on the new computer
  6. Configure Database Credentials:

    • Update environment variables or Cursor MCP configuration with the correct database connection details for the new environment

📁 What's Included vs. What You Need to Build

✅ Included in this repository:

  • Source code (src/ folder)
  • Configuration files (package.json, tsconfig.json, README.md)
  • Documentation files
  • .gitignore (protects against accidentally committing large folders)

🔨 What you need to build after cloning:

  • node_modules/ folder → Run npm install
  • dist/ folder → Run npm run build

⚙️ What you need to customize:

  • Database credentials (host, port, username, password)
  • File paths in Cursor MCP configuration
  • Environment-specific settings

💡 Why this approach?

  • Keeps repository lightweight (MB instead of GB)
  • Avoids platform-specific dependency issues
  • Follows Node.js best practices

Troubleshooting

Setup Issues

  • "Cannot find module" errors: Run npm install to install dependencies
  • "Cannot find dist/index.js": Run npm run build to compile TypeScript
  • Missing node_modules/ folder: This is normal! Run npm install to create it
  • Missing dist/ folder: This is normal! Run npm run build to create it

Connection Issues

  • Verify MySQL server is running
  • Check host, port, username, and password
  • Ensure user has necessary database permissions

Permission Errors

Grant appropriate MySQL privileges:

GRANT SELECT, INSERT, UPDATE, DELETE ON your_database.* TO 'your_user'@'localhost';
FLUSH PRIVILEGES;

Deployment Issues

  • Ensure Node.js is installed on the target computer
  • Verify the file path in Cursor MCP configuration is correct
  • Check that npm install and npm run build completed successfully
  • Confirm database connectivity from the new computer

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选