MCP.so
登录

mpc-csharp-semantickernel

@mrazvanalex

关于 mpc-csharp-semantickernel

Example of usage of Microsoft Semantic Kernel with OpenAI and MCP Server

基本信息

分类

其他

许可证

MIT

运行时

c#

传输方式

stdio

发布者

mrazvanalex

配置

暂无标准配置

该服务器的 README 中没有可解析的 MCP 配置块,请前往代码仓库查看安装说明。

代码仓库

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is mpc-csharp-semantickernel?

mpc-csharp-semantickernel is a sample C# project that demonstrates integration of Microsoft Semantic Kernel with OpenAI’s GPT‑4o-mini and the official C# SDK for the Model Context Protocol. It is intended for developers exploring how to combine Semantic Kernel with MCP and includes a basic email‑sending tool via SMTP. The repository is explicitly marked as not maintained and is meant only as a quick example.

How to use mpc-csharp-semantickernel?

To use the sample: obtain an OpenAI API key; fill out the appsettings for both the WebAppMCPPoc and MCPServer projects (including SMTP server details); update the hardcoded email dictionary in Tools/EmailTool.cs with your own contacts; run the project (the default launch settings point to http://localhost:5109); then send a POST request to https://localhost:7113/chat with a JSON body containing a "text" field to trigger a chat and optionally send an email.

Key features of mpc-csharp-semantickernel

  • Sample integration of Semantic Kernel with OpenAI and MCP C# SDK
  • Sends emails via SMTP using a built‑in tool
  • Email recipient list is hardcoded in a dictionary
  • Uses GPT‑4o-mini as the default chat model
  • Two separate projects: a web client and an MCP server
  • Operates over HTTP on localhost with configurable ports

Use cases of mpc-csharp-semantickernel

  • Learning how to combine Semantic Kernel with the Model Context Protocol
  • Prototyping an AI assistant that can send emails through SMTP
  • Testing OpenAI chat completions within a .NET MCP server
  • Demonstrating a minimal multi‑project MCP setup with tool calling

FAQ from mpc-csharp-semantickernel

What is the difference between this sample and a production‑ready MCP server?

This repository is explicitly not maintained and is only a quick sample. It is not designed for production use; it lacks error handling, security hardening, and flexible configuration.

What dependencies or runtime requirements are needed?

You need an OpenAI API key and access to an SMTP email server. The project is built with .NET and uses the official C# MCP SDK from GitHub. Appsettings must be configured for both the web client and the server.

Where does data (emails and chat history) live?

Email recipients are hardcoded in a dictionary inside Tools/EmailTool.cs. Chat messages are not persisted; they are sent directly to OpenAI and the response is returned to the caller. All data remains local to the running instance.

What are the known limitations of this server?

The repository will not be maintained. Email addresses are hardcoded and must be edited in source code. The sample uses a single SMTP configuration and offers no authentication or authorization beyond the OpenAI API key.

What transports and authentication are supported?

The server communicates over HTTP (no TLS enforcement beyond what is configured in launchSettings). Authentication is limited to the OpenAI API key provided in appsettings. SMTP authentication uses a username and password from the email settings.

评论

其他 分类下的更多 MCP 服务器