McpClientServerExample

McpClientServerExample demonstrates a basic MCP setup with a client and a server that communicate using the Model Context Protocol. The server exposes simple built-in tools that can be invoked by the client, such as Echo (returns the input) and ReverseEcho (returns the reversed input). The client uses the ModelContextProtocol.Client interface to send messages, request GPT-powered responses, and invoke available tools. This example focuses on a minimal, self-contained flow to illustrate client-server interactions and how tools can be discovered and invoked during a session.

Prerequisites:

• .NET 8.0 SDK installed
• OpenAI API key if you plan to use GPT models

1. Clone the repository: git clone <repository-url> cd McpClientServerExample

2. (Optional) Set your OpenAI API key: export OPENAI_API_KEY=your_api_key_here

3. Restore and run the client project: dotnet build dotnet run --project client/McpClient.csproj

4. If you want to run or inspect the server-side separately, ensure you have the corresponding server project setup and run it similarly (dotnet run --project server/McpServer.csproj) and point the client to connect to that server if needed.

Tips:

• Ensure OPENAI_API_KEY is set in your environment if you intend to use GPT models.
• The built-in tools Echo and ReverseEcho are useful for testing basic request/response flows.
• If you modify the client or server, re-build before running to pick up changes.
• Confirm network connectivity between client and server endpoints if they run as separate processes.