Better prompts in four questions Goal, background, limits, done when: four questions that turn a vague request into a prompt that gets a useful first answer.
What is a prompt? A plain explanation A prompt is the text you give an AI tool to tell it what you want. What it can contain and why its wording changes the answer.
Why does the AI give vague answers? Vague answers usually follow vague requests. The AI fills gaps with the most common assumption. How to close the gaps.
How long should a prompt be? A prompt should be as long as it takes to say what you need, and no longer. Why short is not always better, and when to cut.
Why give the AI background information? The AI cannot know who the text is for or what came before. Background closes that gap. What to include and what to leave out.
What does 'done when' mean in a prompt? 'Done when' describes what a good result looks like, so the AI knows when to stop. How to write it with one or two clear criteria.
Does adding an example to a prompt help? Showing the AI one example of what you want is often clearer than describing it. How to use examples and what to watch for.
Should I follow up or start a new chat? When an answer is close but not right, a follow-up is usually faster than a restart. When a fresh start is the better choice.
How do I ask an AI for a good summary? A summary prompt works better when it names the reader, the length and what to keep. A short recipe and common pitfalls.
Does 'act as an expert' help in a prompt? Telling the AI to act as an expert can shift tone, but it does not add knowledge. Naming the audience and the task usually helps more.
#AI #Prompting #ChatGPT #Claude #Productivity
This page gives general information and is not legal, financial or medical advice. [email protected]