@albertsnowden/sql-mcp-server NPM

@albertsnowden/sql-mcp-server

A comprehensive Model Context Protocol (MCP) server that enables Large Language Models to safely interact with SQL databases. This server provides secure database access, schema introspection, and intelligent query execution with built-in safety measures.

Perfect for integration with VS Code MCP extension, Claude Desktop, and other MCP-compatible clients.

🚀 Quick Install

npm install -g @albertsnowden/sql-mcp-server

✨ Features

🔒 Security First: Role-based access control, query validation, and SQL injection prevention
🗃️ Multi-Database Support: PostgreSQL, MySQL, SQLite, and SQL Server
🤖 LLM Optimized: Rich context provision and natural language query support
⚡ High Performance: Connection pooling, query caching, and optimization
📊 Monitoring: Comprehensive logging, metrics, and error tracking
🛡️ Production Ready: Rate limiting, authentication, and audit trails
🔌 VS Code Integration: Seamless integration with VS Code MCP extension

🔧 VS Code Setup

1. Install the MCP Extension

Install the MCP extension for VS Code from the marketplace.

2. Install the SQL Server

npm install -g @albertsnowden/sql-mcp-server

3. Configure VS Code Settings

Add this to your VS Code settings.json:

{
  "mcp.servers": {
    "sql-server": {
      "command": "sql-mcp-server",
      "env": {
        "DB_TYPE": "postgresql",
        "DB_HOST": "localhost",
        "DB_PORT": "5432",
        "DB_NAME": "your_database",
        "DB_USER": "your_user",
        "DB_PASSWORD": "your_password"
      }
    }
  }
}

4. Alternative: Use Configuration File

Create a .env file in your workspace:

DB_TYPE=postgresql
DB_HOST=localhost
DB_PORT=5432
DB_NAME=your_database
DB_USER=your_user
DB_PASSWORD=your_password

📦 Installation Options

Global Installation (Recommended for VS Code)

npm install -g @albertsnowden/sql-mcp-server

Local Project Installation

npm install @albertsnowden/sql-mcp-server

Using npx (No Installation Required)

npx @albertsnowden/sql-mcp-server

🗃️ Database Setup

Create a dedicated user for the MCP server with limited permissions:

PostgreSQL

CREATE USER mcp_user WITH PASSWORD 'secure_password';
GRANT CONNECT ON DATABASE your_database TO mcp_user;
GRANT USAGE ON SCHEMA public TO mcp_user;
GRANT SELECT ON ALL TABLES IN SCHEMA public TO mcp_user;

MySQL

CREATE USER 'mcp_user'@'localhost' IDENTIFIED BY 'secure_password';
GRANT SELECT ON your_database.* TO 'mcp_user'@'localhost';
FLUSH PRIVILEGES;

🛠️ MCP Tools Available

The server exposes these MCP tools for LLM interaction:

Tool	Description	Parameters
`9f1_list_tables`	Get all available tables	`schema?` (optional)
`9f1_describe_table`	Get table schema and relationships	`table_name`
`9f1_execute_query`	Run SELECT queries safely	`query`, `max_rows?`, `timeout?`
`9f1_get_sample_data`	Fetch representative data samples	`table_name`, `limit?`
`9f1_validate_query`	Check query syntax and permissions	`query`
`9f1_search_schema`	Find tables/columns by name	`search_term`

🔐 Environment Variables

Variable	Description	Default	Required
`DB_TYPE`	Database type (postgresql, mysql, sqlite)	postgresql	✅
`DB_HOST`	Database host	localhost	✅
`DB_PORT`	Database port	5432	✅
`DB_NAME`	Database name	-	✅
`DB_USER`	Database user	-	✅
`DB_PASSWORD`	Database password	-	✅
`LOG_LEVEL`	Logging level (error, warn, info, debug)	info	❌
`MAX_QUERY_RESULTS`	Maximum rows returned	10000	❌
`QUERY_TIMEOUT`	Query timeout in seconds	30	❌

💡 Usage Examples

Ask Natural Language Questions

Once configured in VS Code, you can ask your AI assistant questions like:

"Show me all tables in the database"
"What's the structure of the users table?"
"How many orders were placed last month?"
"Find customers who haven't placed orders"

Direct MCP Protocol Usage

{
  "method": "tools/call",
  "params": {
    "name": "9f1_execute_query",
    "arguments": {
      "query": "SELECT COUNT(*) FROM users WHERE created_at > NOW() - INTERVAL '30 days'"
    }
  }
}

🔒 Security Features

Query Validation: Prevents dangerous SQL operations (DROP, ALTER, etc.)
Parameter Binding: All queries use parameterized statements
Rate Limiting: Configurable request limits per client
Audit Logging: All queries and access attempts are logged
Connection Security: SSL/TLS database connections supported
Read-Only by Default: Only SELECT operations allowed unless explicitly configured

🚀 Production Deployment

Docker

FROM node:18-alpine
RUN npm install -g @albertsnowden/sql-mcp-server
CMD ["sql-mcp-server"]

Systemd Service

[Unit]
Description=MCP SQL Server
After=network.target

[Service]
Type=simple
User=mcp
ExecStart=/usr/bin/sql-mcp-server
Environment=DB_TYPE=postgresql
Environment=DB_HOST=localhost
Restart=always

[Install]
WantedBy=multi-user.target

🤝 Contributing

Fork the repository
Create a feature branch: git checkout -b feature/amazing-feature
Make your changes
Run tests: npm test
Commit: git commit -m 'Add amazing feature'
Push: git push origin feature/amazing-feature
Submit a pull request

📋 Changelog

v1.0.0

Initial release
Support for PostgreSQL, MySQL, SQLite
VS Code MCP integration
Comprehensive security features
Production-ready deployment options

📄 License

MIT License - see LICENSE file for details.

🆘 Support

🙏 Acknowledgments

Built with the Model Context Protocol SDK and inspired by the need for secure, intelligent database access for LLMs.

mcp sql llm database model-context-protocol vscode ai postgresql mysql sqlite chatgpt claude

@modelcontextprotocol/sdk @types/pg dotenv mysql2 pg zod

1.0.1

6 months ago

1.0.0

6 months ago