What problem does it solve? Building interoperable AI tool integrations in .NET requires correctly implementing the Model Context Protocol wire format, transports, and capability negotiation, which is error-prone without guidance on the official MCP C# SDK. ## Core Features & Use Cases - Server and Client Construction: Create MCP servers over stdio or Streamable HTTP and connect .NET clients to existing MCP servers using McpClient.CreateAsync. - Capability-Aware Integration: Expose tools, resources, and prompts with [McpServerTool], [McpServerResource], and [McpServerPrompt] attributes, and check ServerCapabilities before using optional features like subscriptions, sampling, or elicitation. - Use Case: A developer wants their ASP.NET Core service to expose internal build tools to an AI agent host. They add ModelContextProtocol.AspNetCore, call AddMcpServer().WithHttpTransport().WithToolsFromAssembly(), map the endpoint with MapMcp("/mcp"), and the tools become discoverable to any MCP client. ## Quick Start Ask the AI to scaffold a minimal MCP stdio server in .NET with an echo tool using the ModelContextProtocol package and attribute-based tool discovery.