A Practical Guide for ASP.NET Developers

What is MCP (Model Context Protocol)?

Simple Architecture View

What is DeepSeek MCP Server?

Why Use MCP with .NET Core?

{ 
  "models": { 
    "deepseek": { 
      "provider": "deepseek", 
      "apiKey": "YOUR_DEEPSEEK_API_KEY", 
      "model": "deepseek-chat" 
    } 
  } 
} 

Step 2: Create an ASP.NET Core Web API

Step 3: Configure HttpClient

builder.Services.AddHttpClient("McpClient", client => 
{ 
    client.BaseAddress = new Uri("http://localhost:3000/"); 
}); 

Step 4: Create a Service Layer (Recommended)

{ 
    Task<string> AskDeepSeekAsync(string prompt); 
} 

{ 
    private readonly HttpClient _httpClient; 
    public McpService(IHttpClientFactory factory) 
    { 
        _httpClient = factory.CreateClient("McpClient"); 
    } 
    public async Task<string> AskDeepSeekAsync(string prompt) 
    { 
        var requestBody = new 
        { 
            model = "deepseek", 
            messages = new[] 
            { 
                new { role = "user", content = prompt } 
            } 
        }; 
        var content = new StringContent( 
            JsonSerializer.Serialize(requestBody), 
            Encoding.UTF8, 
            "application/json" 
        ); 
        var response = await _httpClient.PostAsync( 
            "v1/chat/completions", 
            content 
        ); 
        response.EnsureSuccessStatusCode(); 
        return await response.Content.ReadAsStringAsync(); 
    } 
} 
Register the service: 
builder.Services.AddScoped<IMcpService, McpService>(); 

Step 5: Create the API Controller

[ApiController] 
[Route("api/ai")] 
public class AiController : ControllerBase 
{ 
    private readonly IMcpService _mcpService; 
    public AiController(IMcpService mcpService) 
    { 
        _mcpService = mcpService; 
    } 
    [HttpPost("ask")] 
    public async Task<IActionResult> Ask([FromBody] string prompt) 
    { 
        var result = await _mcpService.AskDeepSeekAsync(prompt); 
        return Ok(result); 
    } 
} 

Step 6: Test the Integration

Response (from DeepSeek via MCP)

{ 
  "answer": "MCP is a protocol that helps AI models communicate with tools and applications in a structured way." 
} 

Best Practices

Real-World Use Cases

Conclusion