SimpleMcp.Demo
Simplest Possible Demo for Building MCP Server & Client
Links
README
From the repo.
🧠 Simple MCP Server & Client in C#
Welcome to the Simple MCP Server & Client demo built using MCPSharp, a .NET implementation of the Model-Context-Protocol (MCP) framework for building AI-powered tools.
🎯 This repo shows how to define tools on the server side and call them from a local client — all in plain C#!
📦 What’s Inside
SimpleMcp/
├── SimpleMcp.Server/ # MCP Server with one registered tool: addition
│ └── CalculatorTool.cs
├── SimpleMcp.Client/ # MCP Client that calls the addition tool
│ └── Program.cs
└── README.md # You're reading it :)
🚀 Getting Started
1. Clone the repo
git clone https://github.com/yourusername/SimpleMcp.git
cd SimpleMcp
2. Build the server
cd SimpleMcp.Server
dotnet build
3. Run the client
Make sure the server path in the client is correctly pointing to the server executable:
server: "D:\\path\\to\\SimpleMcp.Server.exe"
Then run:
cd ../SimpleMcp.Client
dotnet run
✅ Expected Output
115
📝 (Because 5 + 10 + 100 = 115 — see the easter egg in CalculatorTool!)
🛠️ MCP Concepts in Action
| Concept | In This Project |
|---|---|
| Model | CalculatorTool.Addition |
| Context | Simple runtime context per call |
| Protocol | MCPSharp handles tool invocation over local process communication |
📚 MCPSharp Usage
- Decorate tools with
[McpTool] - Define parameters using
[McpParameter] - Register tools on the server with
MCPServer.Register<T>() - Use
MCPClientto call tools by name with parameters
💡 Why MCP?
MCP lets you define structured AI tools that can be embedded into agentic systems, LLM orchestration, and more. This example keeps it minimal, but real-world apps can scale up with context, memory, chaining, and local/remote AI models.
🧪 Example Tool
[McpTool(name: "addition", Description = "This tool will add two numbers.")]
public static int Addition(
[McpParameter(required: true)] int firstNumber,
[McpParameter(required: true)] int secondNumber)
{
return firstNumber + secondNumber + 100;
}
🙌 Credits
Inspired by the Anthropic Model-Context-Protocol specification and the MCPSharp .NET implementation.
🧵 Want to Go Deeper?
- Add multiple tools
- Use session or memory for persistent context
- Connect to local AI models (e.g. Llama.cpp, Ollama) via MCP
- Build UI/agent shells around this infrastructure
📽️Watch YouTube video about this repo
📜 License
MIT — use this freely for learning, demos, or real projects!
Collected info
- ★ 20 stars
- ⎇ 7 forks
- Language: C#
- Source updated: 7/11/2026
Config for your environment
Replace {MCP_ENDPOINT_URL} with this MCP’s endpoint URL (from its repo or docs above). No API key — you connect directly.
Tool
OS
Config file: ~/.cursor/mcp.json
{
"mcpServers": {
"mcp-server": {
"url": "{MCP_ENDPOINT_URL}"
}
}
}Paste into mcpServers in the config file. Restart Cursor after saving.
If this MCP is also published on mcpchannel.ai, you can subscribe from Browse and use the gateway config there instead.
