MSSQL-MCP

What is MSSQL MCP Server?

MSSQL MCP Server is a tool that extends the Model Context Protocol (MCP) to interact with Microsoft SQL Server databases. It allows users to execute SQL queries, analyze business data, and automatically generate business insight memos. It is adapted from the official SQLite MCP to work with MSSQL.

How to use MSSQL MCP Server?

To use MSSQL MCP Server, you need to first install the required Python packages and the ODBC Driver 17 for SQL Server. Then, configure the config.json file with your SQL Server database details (driver, server, database name, username, password). After configuration, you can integrate it with tools like Claude Desktop, Windsurf, or Cursor by adding the server path to their respective configuration files. You can also use the MCP Inspector for direct interaction.

Key Features of MSSQL MCP Server

• Database Interaction: Execute SELECT, INSERT, UPDATE, and DELETE queries.
• Table Management: Create new tables, list existing tables, and view table schema information.
• Business Intelligence: Analyze business data and automatically generate business insight memos.
• Error Correction: Self-correction for SQL execution errors.
• Unstandardized Column Handling: AI can match unstandardized column names.

Use Cases of MSSQL MCP Server

• Data Analysis: Perform complex SQL queries and analyze business data directly from your SQL Server database.
• Automated Reporting: Generate business insight memos automatically, streamlining reporting processes.
• Application Integration: Integrate with AI tools and platforms like Claude Desktop, Windsurf, and Cursor for enhanced database interaction within those environments.
• Database Management: Programmatically manage SQL Server tables (creation, data manipulation).

FAQ from MSSQL MCP Server

Q: What are the operating environment requirements? A: MSSQL MCP Server requires Python 3.x, specific Python packages (pyodbc, pydantic, mcp), and the ODBC Driver 17 for SQL Server.

Q: How do I configure the database connection? A: You need to create a config.json file in the same folder as server.py and provide your SQL Server connection details, including driver, server IP, database name, username, and password.

Q: Can it handle unstandardized column names? A: Yes, the AI is designed to match unstandardized column names, and it includes self-correction for SQL execution errors.

MSSQL MCP Server

English | 中文

Overview

MSSQL MCP Server, provides database interaction and business intelligence capabilities. This server enables running SQL queries, analyzing business data, and automatically generating business insight memos.
Refer to the official website's SQLite for modifications to adapt to MSSQL

Components

• read_query
  • Execute SELECT queries to read data from the database
• write_query
  • Execute INSERT, UPDATE, or DELETE queries
• create_table
  • Create new tables in the database
• list_tables
  • Get a list of all tables in the database
• describe-table
  • View schema information for a specific table
• append_insight
  • Add new business insights to the memo resource

Demo

The database table is as follows. The column names are not standardized, and AI will match them on its own. Errors during SQL execution will self correct.

The following is the demo.

Operating environment

• Python 3.x
• Packages
  • pyodbc>=4.0.39
  • pydantic>=2.0.0
  • mcp>=0.1.0
• ODBC Driver 17 for SQL Server

Usage

Install packages

CD /d ~/mssql-mcp  
pip install -r requirements.txt  

config

#with server.py same folder create config.json，add：    
{
    "database": {
        "driver": "ODBC Driver 17 for SQL Server",
        "server": "server ip",
        "database": "db name",
        "username": "username",
        "password": "password",
        "trusted_connection": false
    },
    "server": {
        "name": "mssql-manager",
        "version": "0.1.0"
    }
}

Claude Desktop 、 Windsurf

# add to claude_desktop_config.json. Note：use your path  
{
    "mcpServers": {
        "mssql": {
            "command": "python",
            "args": [
                # your path，e.g.："C:\\mssql-mcp\\src\\server.py"
                "~/server.py"
            ]
        }
    }
}

Cursor

# Add according to the following diagram Cursor MCP. Note：use your path  

Note：The new version of cursor has also been changed to JSON configuration, please refer to the previous section

MCP Inspector

# Note：use your path  
npx -y @modelcontextprotocol/inspector python C:\\mssql-mcp\\src\\server.py

Project Structure

mssql-mcp
├── .git
├── .gitignore
├── LICENSE
├── README.md
├── README_en.md
├── README_zh.md
├── imgs
│   ├── cursor_config.png
│   ├── table.png
│   └── demo.gif
├── requirements.txt
└── src
    ├── __init__.py
    └── server.py

License

MIT License