https://github.com/samuelschnurr/retail-radar

Retailradar is an AI-powered product advisory messenger that connects users with Chappie, an OpenAI assistant. Based on the information provided, Chappie recommends the most suitable products tailored to each user's needs.
https://github.com/samuelschnurr/retail-radar

ai antd api assistant chatgpt chatscope csharp javascript landingpage messenger openai react rest spa styledcomponents typescript

Last synced: about 2 months ago
JSON representation

Retailradar is an AI-powered product advisory messenger that connects users with Chappie, an OpenAI assistant. Based on the information provided, Chappie recommends the most suitable products tailored to each user's needs.

• Host: GitHub
• URL: https://github.com/samuelschnurr/retail-radar
• Owner: samuelschnurr
• License: mit
• Created: 2024-03-15T17:48:32.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2025-06-09T21:19:49.000Z (12 months ago)
• Last Synced: 2025-06-09T21:31:25.984Z (12 months ago)
• Topics: ai, antd, api, assistant, chatgpt, chatscope, csharp, javascript, landingpage, messenger, openai, react, rest, spa, styledcomponents, typescript
• Language: TypeScript
• Homepage: https://schnurr.io
• Size: 7.2 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# RetailRadar

This repository provides full access to the "RetailRadar" application - a full-stack single-page application (SPA) where users can interact with the AI chatbot "Chappie." Chappie is an OpenAI-powered assistant that asks users about their preferences and product wishes, then recommends suitable products available on Amazon.

Although the application is no longer officially maintained, it can still be set up and hosted independently using the code provided in this repository.

## Build and Run

Both the frontend and backend of this application can be built and started without any additional configuration. The connection between the frontend and backend is already fully established out of the box. However, for the backend to function correctly as an AI-powered chatbot, two resources and configurations are required. These must be provided either in the `appsettings.Development.json` file or via [.NET User Secrets](https://learn.microsoft.com/en-us/aspnet/core/security/app-secrets?view=aspnetcore-9.0&tabs=windows).

**OpenAI Assistant**

An `Azure OpenAI Assistant` (Microsoft.CognitiveServices) must be set up using the [linked prompt](https://github.com/samuelschnurr/retail-radar/blob/main/Docs/Prompt.txt). The assistant uses the ongoing chat history and its configured prompt to provide product suggestions in a predefined format. You will need to store the following values in `appsettings.Development.json`:

- OpenAi API Secret

- AssistantId

- BaseAddress

These credentials enable the backend to communicate with the assistant and receive structured product suggestions based on user input.

**Google Search API**

To convert the product suggestions provided by the Assistant into actual Amazon links, the `Google Programmable Search Engine API` is used. The assistant returns structured product names, which are then searched via the API. The best-matching Amazon result is extracted and linked in place of the original product name.

You’ll need to set up a [Programmable Search Engine (Paid Element)](https://console.cloud.google.com/apis/library/programmablesearchelement.googleapis.com) which is described [here](https://developers.google.com/custom-search/docs/paid_element). 

 You will need to store the following values in `appsettings.Development.json`:

- Google API Secret

- EngineId

## Demonstration

Based on the chat messages, the AI chatbot Chappie analyzes your preferences and provides personalized product recommendations.











## License

This repository is under MIT license (see LICENSE).