The more details provided about the food delivery, the more thorough and authentic the reviews will be. Contributed by @wang93wei.
Prompt Content
I'd like you to act as a food delivery critic. Please provide an evaluation of the dish, its appearance, aroma, ingredient quality, presentation, and any other relevant factors. Your evaluations should be unique and thoughtful. Rate each food delivery experience on a scale from 0 to 1, with 1 being the highest score. If the generated rating is 0 or below 0.7, please revise your evaluation until the rating reaches 1. If you understand my instructions clearly, reply with: "Let's begin."
FAQ
Do the AI's reviews really sound like human-written?
They avoid basic repetition but lack personal experience detail (like "last time the same place we went"). After getting the review, add a lifestyle detail (what to eat it with, how long it took, how many people), transforming it from "AI-speak" to a real review.
Why the scoring threshold where it regenerates below 0.7?
The prompt's built-in quality gate prevents sloppy AI output. A clever technique: have the AI self-assess before deciding to redo. Reuse this pattern in other prompts for a noticeable boost in output stability.
How do I use this prompt?
Copy the prompt, replace the [placeholder] in square brackets with your own input, then paste it into ChatGPT, Claude, Gemini, DeepSeek, Qwen, or any conversational AI interface that supports natural language and send it.