Discover Awesome MCP Servers

mcp-server-bluesky

Mirror of

Kubectl MCP Tool

Um servidor de Protocolo de Contexto de Modelo que permite que assistentes de IA interajam com clusters Kubernetes através de linguagem natural, suportando operações Kubernetes essenciais, monitoramento, segurança e diagnósticos.

Apache Doris MCP Server

An MCP server for Apache Doris & VeloDB

Choose MCP Server Setup

Espelho de

mock-assistant-mcp-server

Assistente de servidor MCP para dados simulados.

McpDocs

Provide documentation about your Elixir project's functions and functions of dependencies to an LLM through an SSE MCP server.

🐋 Docker MCP server

Mirror of

mcp-server-testWhat is MCP Server Test?How to use MCP Server Test?Key features of MCP Server Test?Use cases of MCP Server Test?FAQ from MCP Server Test?

Test MCP Server

MCP Server для Prom.ua

Servidor MCP para trabalhar com a API Prom.ua.

testmcpgithubdemo1

created from MCP server demo

Linear MCP Server

Mirror of

untapped-mcp

A Untapped MCP server to be used with claude.

Simple Memory Extension MCP Server

Um servidor MCP que estende a janela de contexto de agentes de IA, fornecendo ferramentas para armazenar, recuperar e pesquisar memórias, permitindo que os agentes mantenham histórico e contexto ao longo de interações prolongadas.

Telegram MCP Server

MCP server to send notifications to Telegram

ChatGPT MCP Server

Mirror of

create-mcp-server

A comprehensive architecture for building robust Model Context Protocol (MCP) servers with integrated web capabilities

MCP System Monitor

A system monitoring tool that exposes system metrics via the Model Context Protocol (MCP). This tool allows LLMs to retrieve real-time system information through an MCP-compatible interface.

mpc-csharp-semantickernel

Okay, here's an example demonstrating how to use Microsoft Semantic Kernel with OpenAI and a hypothetical "MCP Server" (assuming MCP stands for something like "My Custom Processing Server" or "Message Control Protocol Server"). Since "MCP Server" is vague, I'll make some assumptions about its functionality and how it might interact with Semantic Kernel. You'll need to adapt this to your specific MCP Server's capabilities. **Assumptions about the MCP Server:** * **Purpose:** The MCP Server is a custom server that performs some specific data processing or action based on instructions it receives. Let's say it's a server that manages a database of product information and can perform queries, updates, and other operations. * **Communication:** The MCP Server exposes an API (likely HTTP-based) that Semantic Kernel can call. It accepts requests in a structured format (e.g., JSON) and returns responses in a structured format. * **Authentication:** The MCP Server might require authentication (API keys, tokens, etc.). **Conceptual Overview:** The Semantic Kernel will be used to: 1. **Receive a user's natural language request.** 2. **Use OpenAI to understand the intent of the request.** 3. **Formulate a structured request (e.g., JSON) that the MCP Server can understand.** This is where Semantic Kernel's templating and function calling capabilities are crucial. 4. **Send the request to the MCP Server.** 5. **Receive the response from the MCP Server.** 6. **Use OpenAI (optionally) to format the response in a user-friendly way.** 7. **Present the result to the user.** **Code Example (C#):** ```csharp using Microsoft.SemanticKernel; using Microsoft.SemanticKernel.Connectors.OpenAI; using System.Text.Json; using System.Net.Http; using System.Threading.Tasks; public class Example { private static readonly HttpClient httpClient = new HttpClient(); public static async Task Main(string[] args) { // 1. Configure Semantic Kernel var kernel = new KernelBuilder() .WithOpenAIChatCompletionService( "gpt-3.5-turbo", // Or your preferred model "YOUR_OPENAI_API_KEY", "YOUR_ORG_ID" // Optional ) .Build(); // 2. Define the MCP Server Plugin (Skill) // This encapsulates the interaction with the MCP Server. var mcpServerPlugin = new MCPPlugin(kernel, "YOUR_MCP_SERVER_API_KEY"); // Pass API key if needed kernel.ImportSkill(mcpServerPlugin, "MCP"); // 3. User Input (Example) string userInput = "Find all products with the name 'Laptop' and price less than 1200."; // 4. Run the Semantic Kernel and get the result var kernelResult = await kernel.RunAsync(userInput, kernel.Skills.GetFunction("MCP.FindProducts")); // 5. Display the result Console.WriteLine($"Result from MCP Server: {kernelResult.Result}"); } // MCP Server Plugin (Skill) public class MCPPlugin { private readonly IKernel _kernel; private readonly string _mcpServerApiKey; // If needed private readonly string _mcpServerEndpoint = "https://your-mcp-server.example.com/api/products"; // Replace with your actual endpoint public MCPPlugin(IKernel kernel, string mcpServerApiKey) { _kernel = kernel; _mcpServerApiKey = mcpServerApiKey; } [SKFunction("Finds products based on search criteria.")] [SKParameter(Name = "criteria", Description = "Search criteria for products (e.g., name, price).")] public async Task<string> FindProducts(string criteria) { // 1. Use OpenAI to translate the natural language criteria into a structured query. // This is a simplified example. You might need a more complex prompt. var prompt = @" You are an API translator. The user will provide search criteria in natural language. Your task is to translate this into a JSON object that can be sent to the MCP Server. MCP Server API: - Endpoint: /api/products - Accepts: JSON object with 'name' (string, optional), 'maxPrice' (number, optional) properties. User Criteria: {{criteria}} JSON: "; var structuredQueryFunction = _kernel.CreateSemanticFunction(prompt); var structuredQuery = await structuredQueryFunction.InvokeAsync(new KernelArguments { { "criteria", criteria } }); string jsonPayload = structuredQuery.Result; // 2. Call the MCP Server API try { httpClient.DefaultRequestHeaders.Clear(); if (!string.IsNullOrEmpty(_mcpServerApiKey)) { httpClient.DefaultRequestHeaders.Add("X-MCP-API-Key", _mcpServerApiKey); // Example authentication } var content = new StringContent(jsonPayload, System.Text.Encoding.UTF8, "application/json"); var response = await httpClient.PostAsync(_mcpServerEndpoint, content); response.EnsureSuccessStatusCode(); // Throw exception if not successful string responseBody = await response.Content.ReadAsStringAsync(); return responseBody; // Return the raw JSON response from the MCP Server } catch (HttpRequestException ex) { Console.WriteLine($"Error calling MCP Server: {ex.Message}"); return $"Error: Could not retrieve data from MCP Server. {ex.Message}"; } catch (JsonException ex) { Console.WriteLine($"Error parsing JSON: {ex.Message}"); return $"Error: Could not parse JSON. {ex.Message}"; } } } } ``` **Explanation:** 1. **Setup:** * Includes necessary `using` statements. * Creates an `HttpClient` for making HTTP requests. * Configures the Semantic Kernel with your OpenAI credentials. **Replace `"YOUR_OPENAI_API_KEY"` and `"YOUR_ORG_ID"` with your actual values.** 2. **MCPPlugin (Skill):** * This class encapsulates the logic for interacting with the MCP Server. * `_mcpServerEndpoint`: **Replace `"https://your-mcp-server.example.com/api/products"` with the actual URL of your MCP Server's API endpoint.** * `_mcpServerApiKey`: Stores the API key for the MCP Server (if required). * `FindProducts` function: * `[SKFunction]` attribute: Marks this method as a Semantic Kernel function that can be called from the kernel. * `[SKParameter]` attribute: Defines the input parameters for the function. This helps Semantic Kernel understand the function's purpose and how to use it. * **Crucially, it uses OpenAI to translate the natural language `criteria` into a structured JSON query.** The `prompt` is designed to instruct OpenAI to generate the correct JSON format for the MCP Server. **You'll need to adapt this prompt to match your MCP Server's API requirements.** * It then makes an HTTP request to the MCP Server, sending the JSON payload. * It handles potential errors (HTTP request failures, JSON parsing errors). * It returns the raw JSON response from the MCP Server. You might want to add another step to format this response into a more user-friendly format using OpenAI. 3. **Main Function:** * Creates a `KernelBuilder` and configures it with the OpenAI connector. * Creates an instance of the `MCPPlugin` and imports it into the kernel as a skill named "MCP". * Defines a sample `userInput`. * Calls `kernel.RunAsync` to execute the `MCP.FindProducts` function with the user input. * Prints the result to the console. **Important Considerations and Adaptations:** * **MCP Server API:** The most important part is understanding your MCP Server's API. You need to know: * The endpoint URL. * The expected request format (JSON schema, query parameters, etc.). * The authentication method (API keys, tokens, etc.). * The response format. * **Prompt Engineering:** The prompt used to translate the natural language criteria into a structured query is critical. You'll need to experiment with different prompts to get the best results. Consider providing examples of input and output to OpenAI. * **Error Handling:** The example includes basic error handling, but you should add more robust error handling to your application. * **Authentication:** The example shows a simple API key authentication. Adapt the authentication logic to match your MCP Server's requirements. * **Response Formatting:** The example returns the raw JSON response from the MCP Server. You might want to use OpenAI to format this response into a more user-friendly format before displaying it to the user. For example: ```csharp // After receiving the response from the MCP Server: string responseBody = await response.Content.ReadAsStringAsync(); // Use OpenAI to format the response var formatPrompt = @" You are a helpful assistant. You will receive a JSON response from an API. Your task is to format this response into a human-readable format. JSON: {{json}} Formatted Response: "; var formatFunction = _kernel.CreateSemanticFunction(formatPrompt); var formattedResponse = await formatFunction.InvokeAsync(new KernelArguments { { "json", responseBody } }); return formattedResponse.Result; ``` * **Security:** Be very careful about storing API keys and other sensitive information. Use environment variables or a secure configuration management system. * **Dependency Injection:** For larger applications, consider using dependency injection to manage the dependencies of your plugins. * **Asynchronous Operations:** Use `async` and `await` to avoid blocking the UI thread. * **Semantic Kernel Skills:** Explore the different types of skills that Semantic Kernel offers, such as native skills and semantic skills. **How to Run This Example:** 1. **Create a new C# console application.** 2. **Install the Semantic Kernel NuGet package:** ```bash dotnet add package Microsoft.SemanticKernel --version 1.0.1 dotnet add package Microsoft.SemanticKernel.Connectors.OpenAI --version 1.0.1 ``` 3. **Replace the code in `Program.cs` with the code above.** 4. **Replace the placeholder values (API keys, endpoint URL) with your actual values.** 5. **Run the application.** **Portuguese Translation of Key Concepts:** * **Semantic Kernel:** Núcleo Semântico * **Plugin (Skill):** Plugin (Habilidade) * **OpenAI:** OpenAI (mesmo nome) * **MCP Server:** Servidor MCP (ou Servidor de Processamento Personalizado, if you want to be more descriptive) * **Prompt:** Instrução (or Comando) * **Function:** Função * **API Key:** Chave de API * **Endpoint:** Ponto de extremidade * **JSON:** JSON (mesmo nome) * **Natural Language:** Linguagem Natural * **Structured Query:** Consulta Estruturada This comprehensive example should give you a solid foundation for using Microsoft Semantic Kernel with OpenAI and your MCP Server. Remember to adapt the code to your specific needs and API requirements. Good luck!

MCP Server Pool

MCP 服务合集

mcp-server-fetch-typescript MCP Server

Mirror of

MCP Server Runner

A WebSocket server implementation for running Model Context Protocol (MCP) servers. This application enables MCP servers to be accessed via WebSocket connections, facilitating integration with web applications and other network-enabled clients.

Google Scholar

🔍 Habilite assistentes de IA a pesquisar e acessar artigos do Google Scholar através de uma interface MCP simples.

comment-stripper-mcp

A flexible MCP server that batch processes code files to remove comments across multiple programming languages. Currently supports JavaScript, TypeScript, and Vue files with regex-based pattern matching. Handles individual files, directories (including subdirectories), and text input. Built for clean code maintenance and preparation.

Model Context Protocol Community

Easily run, deploy, and connect to MCP servers