EducationHub.McpSQLServerTools.Sql 1.0.3

EducationHub.McpSQLServerTools.Sql

This is AI MCP Server Tools to Connect AI Agents with Your SQL Server Database

EducationHub.McpSQLServerTools.Sql is a Model Context Protocol (MCP) server tool that enables querying SQL Server databases through MCP.
It provides a simple way to expose SQL query functionality as MCP tools that can be consumed by AI agents or applications.

🚀 Installation

Add the NuGet package to your project:

dotnet add package Microsoft.Extensions.Hosting
dotnet add package EducationHub.McpSQLServerTools.Sql

⚙️ Setup

1. In your .NET project, register the MCP server and load the tools:

using Microsoft.Extensions.DependencyInjection;
using Microsoft.Extensions.Hosting;
using Microsoft.Extensions.Logging;
using ModelContextProtocol.Server;
using EducationHub.McpSQLServerTools.Sql;

var builder = Host.CreateApplicationBuilder(args);

// Enable logging
builder.Logging.AddConsole(consoleLogOptions =>
{
    consoleLogOptions.LogToStandardErrorThreshold = LogLevel.Trace;
});

// Register MCP server with SQL Tools
builder.Services
    .AddMcpServer()
    .WithStdioServerTransport()
    .WithToolsFromAssembly(typeof(SqlQueryTools).Assembly); // load tools

await builder.Build().RunAsync();

🔑 Configure SQL Server Connection

Before running, set your SQL Server connection string in an environment variable:

$env:SQL_SERVER_CONNECTION_STRING="Server=localhost;Database=MyDb;User Id=sa;Password=Your_password123;"

Or in Linux/macOS:

export SQL_SERVER_CONNECTION_STRING="Server=localhost;Database=MyDb;User Id=sa;Password=Your_password123;"

🛠️ Usage

This package provides the tool:

query_table

• Description: Query all rows from a given SQL Server table.
• Input:

  {
    "tableName": "Employees"
  }
  

• Output:
  A list of rows, each represented as a dictionary of column/value pairs.

📖 Example

1. Request

Send this request to the MCP server:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "query_table",
    "arguments": {
      "tableName": "Employees"
    }
  }
}

2. Response

The MCP server will return:

{
  "jsonrpc": "2.0",
  "id": 1,
  "result": {
    "data": [
      {
        "Id": 1,
        "Name": "John Doe",
        "Department": "HR"
      },
      {
        "Id": 2,
        "Name": "Jane Smith",
        "Department": "IT"
      }
    ]
  }
}

🏗️ Example Project

Here’s how to run a test project:

dotnet new console -n McpSqlTest
cd McpSqlTest
dotnet add package EducationHub.McpSQLServerTools.Sql

Replace Program.cs with the setup code above, then run:

dotnet run

⚡ Create mcp.json

To use this MCP server with GitHub Copilot, create a file named mcp.json in your project root folder.
Update the absolute project path inside --project.

Example if your project is located at:
C:\Users\YourName\source\repos\McpSqlTest.csproj

{
  "servers": {
    "sql-server-tools": {
      "command": "dotnet",
      "args": [
        "run",
        "--project",
        "C:/Users/YourName/source/repos/McpSqlTest.csproj"
      ],
      "env": {
        "SQL_SERVER_CONNECTION_STRING": "Server=localhost;Database=MyDb;User Id=sa;Password=Your_password123;"
      }
    }
  }
}

🤖 Testing with GitHub Copilot

1. Open GitHub Copilot Chat in VS Code or GitHub.com.
2. Ensure MCP support is enabled (Copilot Labs → MCP settings).
3. Point Copilot to your mcp.json.
4. Ask Copilot something like:

"Use the sql-server-tools MCP server to query the Employees table."

Copilot will call the query_table tool from your MCP server and return rows directly in chat.

📌 Notes

• Ensure your SQL Server is running and accessible.
• The current implementation supports simple SELECT * FROM <table> queries.
• Future versions will support parameterized queries and filtering.

📦 Publishing

This package is maintained under the EducationHub namespace.
If you want to contribute, fork the repo and submit a PR.