Prompt

A prompt is the instruction you give an AI. Every question, command, or example you type into a chat window is a prompt.

Think of it like ordering at a restaurant. Say "just give me anything" and you'll get whatever comes. Say "less spicy, no cilantro, and make it a large" and you'll get exactly what you want. AI works the same way — the more specific you are about role ("You're a travel guide"), conditions ("budget is $500"), and format ("put it in a table"), the better the result.

The skill of doing this well is called "prompt engineering." It briefly made headlines as a job paying six figures, but at its core it's just "knowing how to give clear instructions." Just like you'd explain the background, goal, and deadline to a new hire, giving AI the right context works the same way.

"An era where a single prompt can build an app" — meaning you can produce a finished product with just instructions (a prompt), no code required.

Try asking the same question two different ways and compare.

1. First: "Recommend a dinner menu."
2. Then: "I have tofu, eggs, and green onions in my fridge. Recommend 3 dinner dishes I can make in 15 minutes, with difficulty levels, in a table."
3. The second answer will be far more useful. Specific conditions plus a clear desired format — those are the two ingredients of a good prompt.

The overlapping names make them sound like two words for the same thing. In reality, what differs is who writes it and when. A prompt is what you type right now; a system prompt is the instruction the AI has already read before you type your first letter. Part of why ChatGPT and Claude sound different isn't just the underlying model — it's this instruction layer.