MCP Microsoft SQL Server
An MCP server for integrating with Microsoft SQL Server databases.
MCP Microsoft SQL Server
A configurable Model Context Protocol (MCP) server for Microsoft SQL Server integration with Claude Code and other MCP clients. Enables AI assistants to securely interact with SQL Server databases through project-based configurations with full read/write capabilities.
🌟 Features
🔧 Project-Based Configuration
- Multiple database connections - Switch between different projects and databases
- Schema-specific access - Restrict access to specific schemas per project
- Configurable permissions - Fine-grained control over read/write/delete operations
- Environment-based setup - Different configurations for dev, staging, and production
🛡️ Security & Safety
- Transaction management - Automatic rollback on errors for write operations
- Query validation - Prevent SQL injection and validate all operations
- WHERE clause enforcement - Mandatory WHERE clauses for UPDATE/DELETE operations
- Row limit restrictions - Configurable limits to prevent accidental mass operations
- Audit logging - Track all database operations for accountability
🔍 Database Operations
- Read Operations: SELECT queries with pagination and filtering
- Write Operations: INSERT, UPDATE, DELETE with transaction safety
- Schema Exploration: Browse tables, columns, relationships, and indexes
- Table Management: Get metadata, statistics, and sample data
- Configuration Management: Switch between project configurations dynamically
🚀 AI Integration
- Claude Desktop integration - Seamless setup with Claude Desktop app
- MCP protocol compliance - Works with any MCP-compatible client
- Natural language interface - Interact with databases using plain English
- Error handling - Clear, actionable error messages for AI and humans
📋 Prerequisites
- .NET 9.0 or later
- Microsoft SQL Server (any supported version)
- Claude Desktop (for Claude integration)
- Appropriate database permissions for the operations you want to perform
🚀 Quick Start
1. Installation
# Clone the repository
git clone https://github.com/yourusername/mcp-ms-sql-server.git
cd mcp-ms-sql-server
# Build the project
dotnet build -c Release
2. Create Project Configuration
Create a configuration file for your project in the Configurations/
directory:
// Configurations/my-project.json
{
"name": "My E-Commerce Project",
"connectionString": "Server=localhost;Database=ECommerceDB;Integrated Security=true;",
"allowedSchema": "dbo",
"permissions": {
"allowRead": true,
"allowWrite": true,
"allowDelete": false
},
"security": {
"requireWhereClause": true,
"maxRowsPerQuery": 1000,
"auditOperations": true
}
}
3. Configure Claude Desktop
Add to your Claude Desktop configuration:
Windows: %APPDATA%\Claude\claude_desktop_config.json
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"sql-server": {
"type": "stdio",
"command": "C:\\Git\\mcp-ms-sql-server\\McpMsSqlServer\\bin\\Release\\net9.0\\McpMsSqlServer.exe",
"env": {
"MCP_CONFIG_NAME": "my-project"
}
}
}
}
Note: First build the project with dotnet build -c Release
to create the executable.
4. Start Using
In Claude Desktop, you can now ask:
- "Show me all tables in the database"
- "Insert a new customer with name 'John Doe' and email 'john@example.com'"
- "What are the top 10 products by sales?"
- "Update the price of product with ID 123 to $29.99"
📖 Configuration Guide
Project Configuration Structure
{
"name": "Project Display Name",
"connectionString": "Your SQL Server connection string",
"allowedSchema": "schema_name",
"permissions": {
"allowRead": true,
"allowWrite": true,
"allowDelete": false,
"allowSchemaChanges": false
},
"security": {
"requireWhereClause": true,
"maxRowsPerQuery": 1000,
"maxRowsPerUpdate": 100,
"maxRowsPerDelete": 10,
"auditOperations": true
},
"querySettings": {
"timeoutSeconds": 30,
"enableQueryPlan": false,
"allowJoins": true,
"allowSubqueries": true
},
"restrictedTables": ["sensitive_table", "audit_log"],
"allowedOperations": ["SELECT", "INSERT", "UPDATE", "DELETE"]
}
Example Configurations
{
"name": "Development Database",
"connectionString": "Server=dev-server;Database=DevDB;Integrated Security=true;",
"allowedSchema": "dbo",
"permissions": {
"allowRead": true,
"allowWrite": true,
"allowDelete": true,
"allowSchemaChanges": true
},
"security": {
"requireWhereClause": false,
"maxRowsPerQuery": 5000,
"auditOperations": false
}
}
{
"name": "Production Database",
"connectionString": "Server=prod-server;Database=ProdDB;User Id=app_user;Password=secure_password;",
"allowedSchema": "app",
"permissions": {
"allowRead": true,
"allowWrite": true,
"allowDelete": false,
"allowSchemaChanges": false
},
"security": {
"requireWhereClause": true,
"maxRowsPerQuery": 100,
"maxRowsPerUpdate": 10,
"auditOperations": true
},
"restrictedTables": ["user_passwords", "payment_info"]
}
🛠️ Available Tools
Configuration Management (4 tools)
- ListConfigurations - Show all available project configurations
- SwitchConfiguration - Switch to a different project configuration
- GetCurrentConfiguration - View current configuration details
- TestConnection - Test database connectivity
Core Database Operations (6 tools)
- ExecuteQuery - Run SELECT queries within allowed schema
- GetSchemaInfo - Explore database structure and objects
- GetTableInfo - Get table metadata and sample data
- InsertRecords - Insert new records with transaction support
- UpdateRecords - Update existing records (requires WHERE clause)
- DeleteRecords - Delete records (requires WHERE clause)
Advanced Features (6 tools)
- BuildQuery - Generate SQL queries from natural language
- AnalyzeQueryPerformance - Analyze query execution plans
- GetDatabasePerformanceStats - Database performance metrics
- DiscoverData - Search tables/columns by patterns
- AnalyzeTableRelationships - Find table relationships
- ProfileDataQuality - Analyze data quality and statistics
🔒 Security Considerations
Best Practices
- Use dedicated database users with minimal required permissions
- Enable audit logging for production environments
- Set appropriate row limits to prevent accidental mass operations
- Restrict sensitive tables using the
restrictedTables
configuration - Use WHERE clause requirements for UPDATE/DELETE operations
- Regular security reviews of configurations and permissions
Connection String Security
# Use environment variables for sensitive data
export DB_PASSWORD="your_secure_password"
{
"connectionString": "Server=myserver;Database=mydb;User Id=myuser;Password=${DB_PASSWORD};"
}
🤝 Contributing
We welcome contributions! Please see our Contributing Guide for details.
Development Setup
# Clone the repo
git clone https://github.com/yourusername/mcp-ms-sql-server.git
cd mcp-ms-sql-server
# Install dependencies
dotnet restore
# Run tests
dotnet test
# Build and test
dotnet build -c Release
📚 Documentation
🐛 Troubleshooting
Common Issues
Connection Failed
# Test your connection string
dotnet run -- --test-connection --config your-project
Permission Denied
- Check your database user permissions
- Verify the
allowedSchema
configuration - Ensure the user has access to the specified schema
Configuration Not Found
- Verify the configuration file exists in
Configurations/
- Check the
MCP_CONFIG_NAME
environment variable - Ensure the JSON syntax is valid
📄 License
This project is licensed under the MIT License - see the LICENSE file for details.
🌟 Acknowledgments
- Model Context Protocol by Anthropic
- MCP C# SDK by Microsoft and Anthropic
- The open-source community for inspiration and feedback
📞 Support
- Issues: GitHub Issues
- Discussions: GitHub Discussions
- Documentation: Wiki
Made with ❤️ for the MCP and AI development community
Related Servers
AI Knowledge System
An enterprise-ready system to archive AI conversations from ChatGPT and Claude into a Supabase database.
Unofficial ChEMBL MCP Server
Access the ChEMBL chemical database for drug discovery, chemical informatics, and bioactivity research using specialized tools via its REST API.
Materials Project MCP
Query the Materials Project database using the mp_api client. Requires an MP_API_KEY environment variable.
DeepMemory
DeepMemory MCP is a small Model Context Protocol (MCP) server that provides long-term memory storage for conversational agents.
MongoDB Atlas
A server for managing data in MongoDB Atlas, providing secure and scalable data management through RESTful APIs.
Hive MCP Server
Enables AI assistants to interact with the Hive blockchain through the Model Context Protocol.
GraphMem
An MCP server for graph-based memory management, enabling AI to create, retrieve, and manage knowledge entities and their relationships.
Biomart MCP
Interface with Biomart, a biological data query tool, using the pybiomart Python package.
Memento MCP
A scalable knowledge graph memory system for LLMs with semantic retrieval and temporal awareness, using Neo4j as a backend.
MCP KQL Server
Execute KQL queries using Azure authentication. Requires Azure CLI login.