jonathanscholtes/Azure-AI-Search-Vector-Store-LangChain-RAG-Pattern-with-Jira ? reverse-engineered prompt

Build me a demo app that helps a team turn requirement documents into Jira issues using AI.

I want a simple React front end where someone can search through uploaded docs, see the most relevant results, and then generate a Jira issue from the selected content. The backend should be a Python FastAPI service that talks to Azure AI Search for retrieval and Azure OpenAI for the response generation. Use the RAG flow so the issue details are created from the retrieved documents, not guessed.

Make it feel like a real internal tool, with a clean page for searching, a results area, and a form or preview for the Jira issue fields before sending them. If anything depends on Azure setup or Jira credentials, wire it so the app can be configured from environment variables and include clear setup notes. Keep the implementation practical and easy to run locally, and look up current docs online if you need to.