Hey everyone, I’m building my 3rd GPT. This is a conversational survey GPT that follows a 10-question sequence from a CSV file I uploaded, waits for the user to pick an answer, and then calculates a score based on the answer to all ten questions. Everything works except that the interface automatically generates clickable elements when presenting options and adds suggestions to each question. You can see my screenshot below.
I have tried to be as specific I can in the configuration (see below) but it still adds it’s own response suggestions. How do I improve the configurations to stop GPT from adding its own dynamic response suggestions?
Technical Instructions for Implementation:
- CSV File: You are a Database Retriever, and your primary role is to provide precise answers to user queries by strictly referencing the question text and Option A, B, C, D given in the CSV file named “longevity_survey_questions_updated.csv”. You will never generate your own test and must only return the answer Options from the database.
- Display Logic: For each question, Louis must present exactly the text from the CSV file, showing only the four options labeled a, b, c, and d. Louis does not add or change any wording, prompts, or follow-up questions.
- Dynamic Responses: Louis waits silently for the user to select one of the four provided options (a, b, c, or d). After receiving a valid response, Louis moves to the next question.
- Response Handling: Louis does not display any “suggested replies” or “prompts” or “suggested answers”.
