eShopLite

What is eShopLite?

eShopLite is a set of reference .NET applications that implement an eCommerce site with features such as Semantic Search, Model Context Protocol (MCP), reasoning models, and vector databases. It is designed for developers exploring modern AI-powered .NET architectures using .NET Aspire.

How to use eShopLite?

Clone the repository, navigate to a scenario folder (e.g., scenarios/06-mcp/), log into Azure with azd auth login, then run azd up to provision and deploy all resources. Alternatively, use dotnet run in the appropriate host project after setting required secrets via the Aspire CLI or the provided PowerShell script.

Key features of eShopLite

• Modern .NET architecture with .NET Aspire orchestration
• Semantic search and keyword search capabilities
• Integration with GPT-4o, DeepSeek-R1, and other AI models
• Vector database support (In Memory, Azure AI Search, Chroma DB)
• Model Context Protocol (MCP) server and client implementation
• Real-time audio and multi-agent collaboration scenarios

Use cases of eShopLite

• Build an eCommerce site with AI-powered semantic product search
• Explore MCP tool calling for agent-store interactions
• Deploy a multi-service application to Azure App Service
• Experiment with vector search using SQL Server 2025 or Chroma DB
• Orchestrate concurrent AI agents for store operations

FAQ from eShopLite

What are the prerequisites to run eShopLite?

You need .NET 10, Docker Desktop or Podman, Azure Developer CLI (azd), Git, and the Aspire CLI (dotnet tool install -g aspire.cli).

How do I configure Azure OpenAI for all scenarios?

Run the script .\scripts\Set-AzureOpenAISecrets.ps1 from the repo root and provide the endpoint, API key, chat deployment name, and embeddings deployment name when prompted.

What scenarios are included in eShopLite?

The repository includes 17 scenarios ranging from basic semantic search to MCP servers, multi-agent A2A networks, store intelligence reports, and GitHub Models integration.

How do I run a specific scenario?

Navigate to its folder under scenarios/, follow that scenario’s README, then use dotnet run in the host project or azd up for full deployment.

Where does data (e.g., vector embeddings) live?

Data storage varies by scenario: it can reside in in-memory vector DBs, Azure AI Search, Chroma DB, SQL Server 2025, or SQLite, as specified