MCP-Server for MSSQL
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP-Server for MSSQLList all tables in the Northwind database"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Server for On-Prem Database Access
This project implements a Model Context Protocol (MCP) server that provides secure access to on-premise MSSQL databases. It allows MCP clients to execute SQL queries, retrieve database metadata, and perform various database operations through a standardized protocol.
Features
Database Tools: Execute SQL queries, stored procedures, and retrieve metadata
Schema Exploration: Get database schemas, tables, views, columns, and distinct values
Secure Authentication: API key-based authentication for database access
Session Management: HTTP-based transport with session handling
Error Monitoring: Integrated Sentry for error tracking and performance monitoring
Docker Containerization: Easy deployment in containerized environments
Related MCP server: mcp-server-mssql
Prerequisites
Node.js 18+
TypeScript
Access to on-premise MSSQL database
Docker and Docker Compose (for containerized deployment)
Installation
Clone the repository
Install dependencies:
npm installBuild the project:
npm run build
Configuration
Environment Variables
Set the following environment variables:
# Database Configuration
DB_USER=your-db-user
DB_PASSWORD=your-db-password
DB_SERVER=your-db-server
DB_NAME=your-db-name
TRUST_CERT=false # Set to true for self-signed certificates
# API Configuration
API_KEY=your-static-api-key
PORT=3000 # Optional, defaults to 3000
# Monitoring
SENTRY_DSN=your-sentry-dsn # Optional, for error trackingDatabase Connection
The server connects to an MSSQL database using the mssql package. Ensure your database server allows connections from the server host.
Usage
Development
npm run devProduction
npm run build
npm startThe server will start on the specified port (default 3000) and listen for MCP requests at /mcp endpoint.
Available Tools
get_databases
Get a list of all databases on the server.
Parameters: None
get_tables
Get a list of all tables in the specified database, organized by schema.
Parameters:
database(string, optional): Database name (defaults to configured default)
get_list_views
Get a list of all views in the specified database, organized by schema.
Parameters:
database(string, optional): Database name (defaults to configured default)
get_columns
Get column information for a specific table or view.
Parameters:
table(string): Fully qualified table name (schema.table)database(string, optional): Database name (defaults to configured default)
run_sql
Execute any SQL command on the database.
Parameters:
query(string): SQL query to execute
execute_stored_procedure
Execute a stored procedure with parameters.
Parameters:
procedure(string): Stored procedure nameparameters(array, optional): Array of parameter objects withnameandvaluedatabase(string, optional): Database name (defaults to configured default)
get_table_joins
Get join information from 'Joins' extended property on tables.
Parameters:
table(string, optional): Specific table name (if not provided, returns joins for all tables)database(string): Database name
get_distinct_values
Get distinct values from a column, capped at 50 unique values, plus the total count.
Parameters:
table(string): Fully qualified table name (schema.table)column(string): Column namedatabase(string, optional): Database name (defaults to configured default)
Docker Deployment
The project includes Docker support for easy deployment.
Prerequisites
Docker and Docker Compose installed
Quick Start
Clone the repository
Configure environment variables in
.envfileStart the services:
docker-compose up --build
This will start the MCP server in a container.
Environment Variables for Docker
# Database
DB_USER=your-db-user
DB_PASSWORD=your-db-password
DB_SERVER=your-db-server
DB_NAME=your-db-name
TRUST_CERT=false
# API
API_KEY=your-api-key
PORT=3000
# Monitoring
SENTRY_DSN=your-sentry-dsnMCP Client Integration
This server can be used with any MCP-compatible client.
Configuration for Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"database-server": {
"command": "node",
"args": ["/path/to/project/dist/index.js"]
}
}
}Testing the Server
You can test the server using curl:
curl -X POST http://localhost:3000/mcp \
-H "Content-Type: application/json" \
-H "Authorization: Bearer your-api-key" \
-d '{"jsonrpc": "2.0", "id": 1, "method": "tools/list", "params": {}}'Security Considerations
Authentication is required for all operations using Bearer token
Database credentials are stored securely as environment variables
All queries are logged for audit purposes
Sensitive data should be handled according to your organization's security policies
Development
Project Structure
src/
index.ts # Main entry point
server.ts # MCP server implementation with HTTP transport
config.ts # Configuration and environment variables
db.ts # Database connection pool management
auth.ts # API key authentication
tools.ts # MCP tool definitions and handlers
dist/ # Compiled JavaScript output
Dockerfile # Docker build configuration
docker-compose.yml # Multi-service Docker setupAdding New Tools
Define the tool schema in the
toolsarray intools.tsImplement the handler function
Add the handler to the
toolHandlersmapUpdate the
handleToolCallfunction if needed
Building and Running
# Development with auto-reload
npm run dev
# Build for production
npm run build
# Run production build
npm start
# Clean build artifacts
npm run cleanMonitoring and Logging
The server integrates with Sentry for error tracking and performance monitoring. Configure SENTRY_DSN to enable.
All database operations are logged to the console for debugging and audit purposes.
This server cannot be deployed
Maintenance
Related MCP Connectors
Remote MCP for 1,500+ APIs. Vault-managed credentials; OAuth or API key. Search, load, and execute.
Governed MCP gateway: one endpoint for your tools, with credential custody and audit log.
Draxlr's remote MCP server connects AI assistants to your SQL databases and dashboards. Explore schemas, run read-only queries, manage saved queries and dashboards, and export results, all with row-level security so each user sees only their own data.
- OneOAuthai.withone
Search, document and execute authenticated API calls across 700+ apps via one MCP server
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables interaction with Microsoft SQL Server databases through T-SQL query execution, table exploration, and schema inspection. Supports configurable write protection and row limiting for safe database operations.628 npmMIT
- AlicenseBqualityCmaintenanceEnables AI agents to interact with Microsoft SQL Server databases via MCP, supporting table listing, schema retrieval, and CRUD operations.31MIT
- AlicenseNot gradedqualityCmaintenanceProvides tools to connect, query, and manage Microsoft SQL Server databases through the MCP protocol, including support for stored procedures, transactions, and schema inspection.72 PyPIMIT
- AlicenseNot gradedqualityAmaintenanceEnables secure, controlled access to existing relational databases via MCP, with API key authorization, read/write permissions, table-level allow/deny lists, SQL validation, rate limiting, audit logging, and a web management console. Supports PostgreSQL, MySQL, SQLite, SQL Server, and Oracle.9 npm1MIT