🔍
Press Esc to close • Ctrl+K to open

Effective AI Prompt Writing Using the Pentagon Model

In this course you will move from random requests to professional prompts through five connected pillars: role, context, data, task & output, and constraints.

Lessons: 8 Completed: 0/8 Path: Progressive
🧪 Hands-on lab, in-page
🤖 Try prompts on a real AI
🎓 Shareable certificate
🎯 Focused, no fluff
Pentagon Model

Refine Your Prompt: from good to professional

A practical course for building clear and powerful AI prompts

🎉

Well done! You completed the course

You mastered all five pillars and applied them hands-on. Your certificate is ready — claim it and share it.

🎓 Get your certificate

You have mastered the five pillars. But in reality, a prompt is rarely perfect on the first try — the real skill is not writing a flawless prompt at once, but reading the output and refining the prompt. This is a short loop you repeat until you reach the result you want.

Core principle
An effective prompt is built by iteration, not instant perfection. Run it, read the output critically, spot the weakest pillar, fix only that, and re-run.

The refine loop in 4 steps

  • Run your prompt and read the full output.
  • Ask: where did it fall short? (tone, missing detail, messy shape, wrong assumption)
  • Map the gap to its pillar: tone → role/constraints, detail → data, order → task & outputs.
  • Edit only that pillar and re-run — do not rewrite from scratch.
First draft
Write a marketing message for a perfume store.
After two refines
Act as a marketing expert. For a premium perfume store targeting women before Eid, write a 3-line, elegant, ready-to-send message.
Pro tip
The lab in the next lesson is built for this: change one pillar, re-run, and watch the output shift. That is the refine loop in practice.
🤖

Ask the Coach

An AI assistant that answers based on this lesson. Ask freely.

🤖 Coach answers are AI-generated; verify important details.

Check Your Understanding (2 questions)

Question 1

You read your output and found the tax calculation field marked 'not available in the quote', although the rate and the amount are written in your prompt. Which pillar do you change?

💡 Why: Data is not the fault: the amount 620,800 and the rate 15% are in fact written in your prompt, and copying them again changes nothing. Output is not the fault: the field was requested by number, and the order came out as you asked. The fault is that your Constraint banned writing 'a price or a rate' in general terms, and with it banned the calculation you had asked for yourself. So narrow the Constraint to its purpose; do not delete it.
Question 2

You obtained the soil-test price from an accredited laboratory. It is now a number you hold, and you want the output to calculate with it. Where do you put it?

💡 Why: Constraints govern what is allowed and what is forbidden. They do not carry your numbers, and if you permitted estimation there, back comes the invented number that the Constraints fixed in the first place. Output is shape, not material: a new column stays empty unless you give the number. The number is raw material, so its place is Data, with its unit and its source.
The recommended next step unlocks only after the correct answer, and your progress is saved on this device.