mssqlMCP
Verified Safeby MCPRUNNER
Overview
The SQL Server MCP (Model Context Protocol) Server enables AI assistants like GitHub Copilot within Visual Studio Code to interact with SQL Server databases, allowing for SQL query execution, detailed database metadata retrieval, and management of connections and security.
Installation
docker run -d --name mssqlmcp -p 3001:3001 -e MSSQL_MCP_KEY="StrongEncryptionKeyForEncryptionOfYourConnectionStrings" -e MSSQL_MCP_API_KEY="StrongApiKeyYourConnections" -v "$(pwd)/data:/app/Data" -v "$(pwd)/logs:/app/Logs" mcprunner/mssqlmcp;Environment Variables
- MSSQL_MCP_KEY
- MSSQL_MCP_API_KEY
- MSSQL_MCP_DATA
- ASPNETCORE_ENVIRONMENT
- ASPNETCORE_URLS
- MSSQL_MCP_TRANSPORT
Security Notes
The project implements robust security features, including AES-256 encryption for connection strings, multi-tier API key authentication with role-based access control and connection-level restrictions, and comprehensive input validation to prevent SQL injection and dangerous commands. Encryption keys and API keys are managed via environment variables and scripts to avoid hardcoding. While the server defaults to HTTP, the documentation explicitly recommends HTTPS for production environments, which would require an external reverse proxy. Regular key rotation and monitoring are also advised for optimal security.
Similar Servers
mcp
Provides a Model Context Protocol (MCP) server for Neo4j, enabling LLMs to interact with the database via defined tools.
MsSqlMCP
Provides an MCP Server for SQL Server database schema inspection and safe, read-only query execution to AI clients.
warp-sql-server-mcp
Enables AI assistants like GitHub Copilot and Warp AI to securely interact with Microsoft SQL Server databases using natural language queries, providing enterprise-grade security, performance monitoring, and data management capabilities.
dbt-core-mcp
Enhances AI assistants (like Copilot) to intelligently interact with dbt projects by executing commands and providing project metadata, lineage, and compiled SQL, respecting the user's local environment.