Build useful prompts one layer at a time

This course remembers completed sections, assessment point totals, and your last course page in this browser on this device. RayAI does not receive this progress or anything you write. Selected answers and written responses are not saved. Clearing browser data, using private browsing, or changing browser or device can remove or hide saved progress.

MODULE 2 | 40 minutes | FAST TRACK CORE + GO DEEPER

Build useful prompts one layer at a time

Core idea: start small, strengthen deliberately

The RayAI framework has seven parts, introduced here in three layers. No earlier framework is required. Use the first layer for ordinary tasks and add the others when ambiguity, consequence, or repeatability increases.

Layer 1 - Start

• Task: What exact action should the model perform? Use a concrete verb: summarize, compare, draft, classify, extract, explain, critique, or propose.

• Context: What facts or source material are necessary? Include only information that could change the answer.

• Output: What form makes the result usable? Specify length, structure, fields, audience level, or tone only when they matter.

START TEMPLATE

Layer 2 - Strengthen

• Audience: Who will use the result, and what do they already know?

• Boundaries: What must the model not assume, disclose, include, or do? What should happen if a critical fact is missing?

Layer 3 - Trust

• Purpose: What outcome or decision will this support? Who owns that decision?

• Quality check: What should be verified, cited, labeled uncertain, recalculated, or escalated before use?

See it: one prompt grows in three layers

Fictional planning notes: An event for adult community members has a $600 budget ceiling. Maya offered to contact two possible speakers. The date and venue are not confirmed.

START

STRENGTHEN

TRUST

Roles help with perspective, not authority

A role such as 'act as an editor' can guide vocabulary or perspective. It does not make the model a licensed professional, a policy owner, or an accountable decision-maker. Prefer a precise task and audience over grand titles such as 'world's best expert.'

Fix it

BLOATED PROMPT

Replace the hype and contradictions with a specific Task, relevant Context, usable Output, and one Quality check.

Try it: practice the skill

What to do

1. Choose the task suggested for this module, or use another simple, low-risk task.
2. Open ChatGPT or another AI assistant in a separate tab or window.
3. For this module, rewrite your prompt using Task + Context + Output. Add only the details that make the answer more useful.
4. Enter your prompt and send it. Do not include private or sensitive information.
5. Read the AI answer.
6. Return to this course and complete the short notes below.
   If you cannot use an AI assistant now, read the worked example and write what you would try later. Skip the steps that ask you to open, send or rerun a prompt. Label your notes "supplied example" or "planned test". Describe only what the example shows; if no output is supplied, write what you would look for. Keep checks you performed separate from checks you plan. Do not invent an AI result.