
You do not need all six parts every time — but checking all six before you hit send catches most of the gaps that cause weak answers.
Why a framework beats a list of prompts
Saved prompts break when your situation changes. A framework adapts: it reminds you what information the model needs, whatever the task. That is the difference between collecting prompts and knowing how to prompt.
The six parts
- Role — the expertise and standards to apply (“an experienced HR manager reviewing a policy for clarity”).
- Task — one clear verb and outcome (“rewrite”, “compare”, “plan”).
- Context — background, audience, what has been tried and why it matters.
- Format — the shape of the answer: table, bullets, sections, word count.
- Constraints — tone, reading level, must-include points, things to avoid.
- Examples — one or two samples of what good looks like.
A full example
Role: You are a senior email marketer for small online shops. Task: Write a 3-email welcome sequence. Context: I sell handmade candles to women aged 25–45; new subscribers get a 10% code. Most buy as gifts. Format: For each email: subject line, preview text, body under 120 words, one call to action. Constraints: Warm, not pushy. No exclamation marks in subject lines. Mention gift wrapping in email 2. Example tone: “Light something that smells like the weekend.”

AI Mastery Bundle
The next step after the basics: frameworks, precision prompting and a 1000+ prompt library.
$50 (was $100) · Look insideWhen to use the short version
For quick questions, task plus context is often enough. Add format when you need to reuse the output, and constraints and examples when tone or accuracy matter.
Make it reusable
Put role, standing context and constraints into custom instructions, a Claude Project or a Gemini Skill. Then each new prompt only needs the task, specifics and format.
Our Prompt Builder walks through all six parts and assembles the prompt for you.
Put it into practice
Open the free Prompt Builder: build a complete prompt with the 6-part framework. This article is part of The Complete ChatGPT Guide (2026 Edition).
Sources & further reading
- OpenAI: Prompt engineering guide
- Anthropic: Prompt engineering overview
- Google: Prompt design strategies
Product facts were checked against official sources on 16 October 2026. AI tools change quickly, so check the provider’s current documentation for anything you rely on. Our editorial policy.


