ASP.NET Core MCP Server - Complete Setup Guide

Prerequisites

• An ASP.NET Core API project targeting .NET 6 or later.
• Swashbuckle.AspNetCore installed and configured.
• Access to LegacyAI for MCP generation.

Step 1: Add Swashbuckle

Swashbuckle generates your OpenAPI document at runtime. Install the package and register Swagger services.

dotnet add package Swashbuckle.AspNetCore

Step 2: Configure Swagger in Program.cs

Enable Swagger for your API and set the OpenAPI metadata. This ensures the spec includes the correct title, version, and auth schemes.

var builder = WebApplication.CreateBuilder(args);

builder.Services.AddControllers();
builder.Services.AddEndpointsApiExplorer();
builder.Services.AddSwaggerGen(options =>
{
    options.SwaggerDoc("v1", new() { Title = "LegacyAI API", Version = "v1" });
});

var app = builder.Build();

app.UseSwagger();
app.UseSwaggerUI();

app.MapControllers();
app.Run();

Step 3: Export the OpenAPI spec

Run the API locally and fetch the JSON from /swagger/v1/swagger.json. Save it as openapi.json for LegacyAI.

curl http://localhost:5000/swagger/v1/swagger.json -o openapi.json

Step 4: Generate MCP tools with LegacyAI

Upload the OpenAPI file to LegacyAI. The generator will map controllers and routes to MCP tools, apply validation, and detect auth schemes.

Use the dashboard to enable read-only mode and configure tool scopes before publishing.

Step 5: Connect Claude Desktop

Add your MCP server to Claude Desktop using the MCP configuration file. Make sure you store tokens in environment variables and not in plain text.

{
  "mcpServers": [
    {
      "name": "LegacyAI ASP.NET Core",
      "command": "node",
      "args": ["./legacyai-mcp-server.js"],
      "env": {
        "API_BASE_URL": "https://api.yourcompany.com",
        "API_TOKEN": "${API_TOKEN}"
      }
    }
  ]
}

Step 6: Verify tool calls

In Claude Desktop, ask the agent to call a read-only endpoint and verify that it returns expected data. Review logs for validation errors or missing headers.