CraftPrompts

ChatGPT prompt generator

Builds prompts in the sectioned markdown style OpenAI’s own prompting guide recommends: clear headings, explicit output rules, and an invitation to ask before guessing.

Start from

Your prompt
roletaskcontextformattonerulesexample
Fill in a few fields and your prompt is set here, line by line.
0 words · 0 characters

Prompt score

Underspecified

0 words

  • Clear task verb missingStart with an explicit action: “Write…”, “Summarize…”, “Compare…”.
  • Background or context missingSay who it is for and why you need it; models fill gaps with generic guesses.
  • Role or perspective missingGive a perspective: “You are a senior UX writer…”. It sets vocabulary and depth.
  • Output format stated missingName the shape of the answer: table, bullets, JSON, 3 paragraphs.
  • Length or scope limit missingCap the length: “under 150 words”, “5 bullets”.
  • Constraints or rules missingAdd rules: what to avoid, what must be included, reading level.
  • Example or reference missingShow one example of the output you want. Few-shot beats description.
  • Specific, not vague missingReplace vague words (good, nice, interesting) with concrete criteria.

Let AI polish it

Sends your filled-in fields to the AI helper, which writes a complete prompt and marks gaps as [placeholders].

What makes ChatGPT prompts work

OpenAI’s prompt engineering guide boils down to a handful of habits: write clear instructions, give reference text, split complex tasks into steps, and give the model room to think. The generator covers the first two with its fields, and the chain-of-thought guide covers the rest.

Headings matter more than they look. When a prompt mixes instructions with pasted material, ChatGPT can treat part of your material as instructions. Putting pasted text under its own heading (or inside triple quotes) keeps the two apart. Looking for ready-made ideas instead? Browse the ChatGPT prompt templates.

Questions

How is a ChatGPT prompt different from a Claude prompt?

Both models follow clear, structured instructions. OpenAI’s guidance favours markdown headings and delimiters such as ### or triple quotes, so this generator uses ### Role, ### Task and so on. Claude responds especially well to XML tags; use the Claude prompt generator for that format.

Should I put this in Custom Instructions or the chat?

Put one-off tasks in the chat. Put stable preferences (who you are, how you like answers formatted) in Custom Instructions or a custom GPT’s instructions. The System Prompt Generator builds that second kind.

Why does the prompt end by asking for questions?

Inviting up to three clarifying questions stops ChatGPT from filling gaps with generic assumptions. Delete the line if you want an answer straight away.

Does it work with GPT-5 and newer models?

Yes. Newer models need less trickery, not less clarity. A precise task, context and format still produce the biggest quality jump.

More tools in the studio