Free preview6 min

The 5-Component Framework

Reviewed by Human · Updated July 22, 2026

Here's the myth: great prompts are long. Detailed. Exhaustive. The truth is stranger — the best prompts in production are complete in exactly the right places and deliberately empty everywhere else.

 

This module hands you the x-ray glasses. Once you can see the skeleton under a prompt, you can't unsee it — and every prompt you read from now on, including bad ones, becomes a free lesson.

Example

Here's a prompt someone might actually send before a family trip. Read it once normally. Then look at the color-coded breakdown — each sentence has exactly one job.

Prompt

You are a senior travel editor. I'm planning a 4-day trip to Kyoto in November with my parents, who walk slowly and love food. Plan a day-by-day itinerary. Format it as a table with columns for morning, afternoon, and evening. Keep walking under 5 km per day and skip anything that needs advance reservations.

AI Response

| Day | Morning | Afternoon | Evening | |---|---|---|---| | 1 | Nishiki Market food walk (flat, ~1.2 km) | Tea and sweets in a Gion teahouse | Pontocho Alley dinner, riverside stroll | | 2 | Taxi to Kinkaku-ji, short loop path | Lunch near Ryoan-ji, rock garden bench time | Early kaiseki-style dinner downtown | | 3 | Philosopher's Path, southern half only (~1.5 km) | Nanzen-ji grounds, café break | Kyoto Station food floor — zero walking, big payoff | | 4 | Fushimi Inari lower gates only (~1 km) | Sake district tasting rooms | Farewell dinner near hotel |
Five sentences, five jobs, nothing decorative. The response respects every component: the table shape, the walking cap, the food emphasis, the no-reservation rule.

Now the twist. If five components make prompts great, expert prompt writers must fill all five every time... right?

How Do You Structure a Prompt?

**Prompt**: A text instruction you send to an AI model to get a response. Every strong prompt draws from five components: **Role**: Who the AI should act as while answering. **Context**: The background facts the AI can't guess — your situation, your audience, your raw material. **Task**: The specific action you want performed. The one verb the whole prompt orbits. **Format**: The shape of the output — table, list, tone, length. **Constraint**: The boundaries — what to avoid, limits to respect, rules that must hold. Treat them as slots, not blanks. A form demands every blank get filled. A mental model asks a sharper question: which slots earn their place for this task? One early answer worth keeping: context is everything the model can't infer on its own — it's the slot beginners skip most and experts fill first.

Memorize the order as R-C-T-F-C: Role, Context, Task, Format, Constraints. You'll map prompts against it for the rest of this course.

Your turn

Your turn to dissect. Below is a working prompt. Label each sentence with its component — R (role), C (context), T (task), F (format), or K (constraint).

Reflect

Which sentence could you delete with the least damage to the output? Hold that question — it's the heart of Module 4.

📚

Go deeper

end of module

You've finished this module.

Mark it complete to earn your XP and keep your streak alive.

Progress saved locally · Sign up to earn XP

When to Use Each Component→