Prompt

A prompt is the input you give a generative AI model, usually natural-language text but sometimes images or files, that describes the task and any context. The model's response is a continuation shaped by that input.

Why it matters

Output quality depends heavily on prompt clarity. Specific instructions, relevant context, and a stated format reduce vague or off-target answers. In applications, prompts are effectively part of your code and deserve version control and testing.

Example

Vague prompt: "Tell me about DNS."

Clearer prompt:

Explain DNS to a non-technical small business owner in
three short paragraphs. End with one practical tip.

Many APIs also separate a system prompt (standing rules) from the user prompt (the current request).

Common mistakes

• Leaving out context the model cannot know, such as your audience or constraints.
• Packing several unrelated tasks into one prompt.
• Putting secrets or private data in prompts that are logged or sent to third parties.

Related

• Prompt engineering
• Context window
• Token
• Large language model

This page is general information written by our editorial team and updated when documentation changes. Servers and providers differ, so test changes safely first. See our editorial policy and disclaimer.