Practical overview
When it fits
Use it when you want a self-hosted reference for assembling agent orchestration, tool integrations and semantic search in an e-commerce application.
Who it is for
- Developers studying multi-agent orchestration patterns in an e-commerce application.
- Teams exploring MCP tool calls and A2A communication with external agents.
- C# developers examining semantic search with Qdrant and Semantic Kernel.
Documented uses
- Explore parent-child, GroupChat and Sequential agent orchestration patterns.
- See how agents call MCP server endpoints and third-party tools over HTTP and JSON-RPC.
- Study external-agent communication using A2A over HTTP and JSON-RPC.
- Examine vector-based semantic and hybrid search for product discovery.
Requirements & limits
The repository is public and MIT-licensed. Running the application requires a self-hosted development environment; the README lists .NET 10.0 SDK, Docker and Aspire CLI.
- The project is a C#/.NET development application rather than a hosted e-commerce service.
- It uses Qdrant for vector storage and supports semantic and hybrid search.
- Chat completion and embedding generation have separate endpoint settings.
- The README identifies a separate Microsoft Agent Framework repository as the migrated implementation; it does not give a direct migration procedure here.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Configure local model-provider endpoints
Set the documented environment configuration for chat and embedding providers.
Before you start
- .NET 10.0 SDK
- A provider exposing an OpenAI-compatible /v1 API shape
- Provider base URL, model names and API keys
Steps
Copy .env.example to .env.
Set SemanticKernelOptions__ChatEndpoint and SemanticKernelOptions__EmbeddingEndpoint to the provider base URL.
Replace the placeholder model names and API keys in the local configuration.
Expected outcome
The application has local configuration for its chat and embedding endpoints.
Guide limitations
- The README describes configuration for local testing; this guide does not cover running the application.
- Keep .env local and do not commit credentials.
Start the application with Aspire
Launch the application through the Aspire App Host in development mode.
Before you start
- .NET 10.0 SDK
- Docker
- Aspire CLI
- A configured model provider and local environment
Steps
Install the Aspire CLI with dotnet tool install -g Aspire.Cli.
Run aspire run from the project environment.
Open the Aspire dashboard at https://localhost:17056 or http://localhost:15234.
Expected outcome
The application is started through Aspire and its development dashboard is available at the documented local addresses.
Guide limitations
- The repository presents Aspire as a development-mode run path.
- The documented dashboard addresses are local to the running development environment.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 15 October 2026
- Original website title
- mehdihadeli/genai-eshop-semantic-kernel · Official homepage title · retained in its source language