RAG Chat Test
This feature helps the administrator test RAG (Retrieval-Augmented Generation) queries against the repository. Each question is analyzed, the relevant documents are retrieved from the vector store, and the AI generates an answer based on that retrieved context.
This menu entry only appears under Administration > Utilities > AI when the chatbot (RAG) feature is enabled.
Check RAG Chat
- Write a question in the Question field.
- Press Enter to send the question, or click the Send button. Use Shift+Enter to add a new line without sending.
- Click the Clear button to reset the question and the response.
The Response panel shows the answer along with the list of Nodes used, with links to the documents retrieved as context to compose it.
Unlike AI Chat test, each question here is processed independently: no conversation history is kept between questions.
