Build better prompts
Prompt Design Framework
The Prompt Design Framework structures a prompt in eight parts: role, objective, context, input, constraints, process, output format and quality criteria, so that results are consistent and easy to check.
The model
- 1Role
- 2Objective
- 3Context
- 4Input
- 5Constraints
- 6Process
- 7Output format
- 8Quality criteria
Developed by Markaigen · Updated
What each part means
- Role
- Who the model should act as, and for whom: it sets expertise and tone.
- Objective
- The one result you want from this prompt, stated as an outcome.
- Context
- What the model needs to know about the brand, audience, channel and situation.
- Input
- The material to work with: text, data, examples, or a placeholder where you will paste it.
- Constraints
- What it must not do: claims to avoid, length, words, formats, legal limits.
- Process
- The steps to follow, in order, including when to ask a question instead of guessing.
- Output format
- Exactly how the result should look: structure, headings, fields, length.
- Quality criteria
- How the model should check its own result before answering.
How to use it
- Write constraints and quality criteria first; they prevent most bad output.
- Keep one prompt per task and save the ones that work as team templates.
- When the result is wrong, change one part at a time, so you learn which part caused it.
Build a prompt
Fill in the eight parts. The page assembles the prompt; copy it into any AI assistant.
Answers are saved in this browser so you can return to them.
Download or e-mail this framework
Get a branded PDF or Word document, with your answers or as an empty template to fill in.