eShopLite

eShopLite is a set of reference .NET applications implementing an eCommerce site with features like Semantic Search, MCP, Reasoning models and more.

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

View docs

Command

claude mcp add --transport stdio azure-samples-eshoplite dotnet run --project path/to/eshoplite/scenarios/06-mcp/McpServer/McpServer.csproj \
  --env MCP_PORT="Port for MCP server to listen on (e.g., 5000)" \
  --env MCP_LOG_LEVEL="Logging level (Information|Debug|Warning|Error)"

How to use

eShopLite includes a dedicated MCP server as part of its scenarios, demonstrating the Model Context Protocol machinery in a realistic eCommerce-like setting. The MCP server enables structured, context-rich interactions between an MCP client and server, including model reasoning, function calling, and event streaming via Server-Sent Events. Use this MCP implementation to explore multi-turn AI reasoning, passing fragmented context, and coordinating between vector stores and LLMs within the eShopLite ecosystem. Start the MCP server first, then connect an MCP client to exercise semantic and tool-using conversations, inspect the context sharing, and observe how model calls are orchestrated through the MCP endpoints.

How to install

Prerequisites:

.NET 9 SDK installed
Git
A compatible environment for running the eShopLite repository (Windows, macOS, or Linux)

Installation steps:

Clone the repository:

git clone https://github.com/azure-samples/eshoplite.git

Navigate to the MCP scenario folder:

cd eshoplite/scenarios/06-mcp

Restore and run the MCP server project:

dotnet restore
dotnet run --project McpServer/McpServer.csproj

Ensure prerequisites are running (e.g., any dependent vector stores or AI services) and note the port printed by the server for client connections.

Additional notes

The MCP server integrates with the eShopLite MCP client to demonstrate model-context-based reasoning.
If the server fails to bind to the default port, set MCP_PORT via environment variable or adjust the port in the project settings.
Common issues include mismatched API versions between client and server, and missing credentials for any external vector databases or AI services.
Environment variables you may encounter or adjust: MCP_PORT, MCP_LOG_LEVEL, VECTOR_DB_CONNECTION_STRING, OPENAI_API_KEY.
To test, use an MCP client wired to the same port and experiment with semantic prompts and function-calling flows as shown in the MCP scenario.

Related MCP Servers

vestige

398

Cognitive memory for AI agents — FSRS-6 spaced repetition, 29 brain modules, 3D dashboard, single 22MB Rust binary. MCP server for Claude, Cursor, VS Code, Xcode, JetBrains.

mcp-agent-langchainjs

169

Serverless AI agent using LangChain.js and Model Context Protocol (MCP) integration to order burgers from a burger restaurant

context-harness

Local-first context ingestion and retrieval for AI tools. SQLite + embeddings + MCP server for Cursor & Claude.

rag

⚠️ DEPRECATED - Use https://github.com/SylphxAI/coderag instead

srag

Semantic code search and RAG system written in Rust with tree-sitter chunking, MCP server for IDE integration, prompt injection detection, and secret redaction

mcp-ragex

MCP server for intelligent code search: semantic (RAG), symbolic (tree-sitter), and regex (ripgrep) search modes. Built for Claude Code and AI coding assistants.