Prompting for research, writing, planning, and learning

A useful prompt states the outcome, audience, context, constraints, evidence standard, and output format. Iteration matters more than finding one magical phrase.

At a glance

Question	Practical answer
When is it useful?	Instead of “research databases,” ask for a beginner comparison for a small shop, require assumptions and primary sources, then request a decision table and unanswered questions.
What should you do?	Rewrite one vague request with six fields: goal, reader, inputs, boundaries, quality checks, and desired format; compare both outputs.
How do you know it worked?	The improved response is easier to verify, contains fewer unsupported assumptions, and can be reused as a concrete next action.
Common failure	Do not ask the model to hide uncertainty; require it to separate facts, assumptions, and recommendations.

flowchart LR
  A[Question] --> B[Prompting for research, writing, planning,]
  B --> C[Small example]
  C --> D[Evidence]

The important idea is not to stop at a definition: connect the concept to a small example and observable evidence.

Worked example

Instead of “research databases,” ask for a beginner comparison for a small shop, require assumptions and primary sources, then request a decision table and unanswered questions.

Before acting, write the success signal. Change one condition at a time, observe the result, and record assumptions. For Prompting for research, writing, planning, and learning, this separates what you know from what you are merely guessing.

Practice in 20–30 minutes

Goal: Rewrite one vague request with six fields: goal, reader, inputs, boundaries, quality checks, and desired format; compare both outputs.

1. Record the starting state and your prediction.
2. Implement the smallest version without adding unnecessary tools.
3. Change exactly one input or constraint and repeat.
4. Save a command, screenshot, output, or checklist as evidence.

Expected result: The improved response is easier to verify, contains fewer unsupported assumptions, and can be reused as a concrete next action.

What can go wrong

Do not ask the model to hide uncertainty; require it to separate facts, assumptions, and recommendations.

When the result differs from your prediction, do not change many things at once. Check inputs, versions, environment, permissions, and logs, then repeat from the smallest example.

Definition of done

• I can explain the concept in my own words.
• I completed the small example and kept evidence.
• I know one failure mode and how to check it.
• Someone else can repeat the work without guessing missing steps.

Go deeper

Use the linked resource or repository at the end of the page when you need a full implementation. Check current versions before applying commands to a real project.