Learn AI. Do. Grow.
← All modules

Deep dive · module 52 of 100

Organising a prompt library

Save good prompts so that you can reuse and share them.

After this lesson

You create a simple structure to name, save, and find prompts.

In view · How it connects

A library you can find again

  1. Give every prompt a recognisable task name and group related tasks.

  2. Save when the prompt works and what you need to fill in.

  3. Make it clear who updates and reviews the shared prompt.

Remember: A found prompt is only useful if the context is included. Open het volledige schema ↗

Give every prompt a name

For example: “Follow-up email after meeting – short and friendly”.

Save context

Note down for which situation the prompt works and what needs to be filled in.

Share with the team

Put good prompts in a shared location so others can use them too.

New practice example

From explanation to a usable result.

A prompt library stores not only a text, but also what it works for and how you check it.

Fictional practice situation

A fictitious team has twenty prompts with names like test, new, and final. No one knows which one is intended for a customer email with an unknown delivery date.

View a possible solution

This is how it can be done

Create a card “Customer email · delivery unknown” with objective, required source fields, fillable prompt, good example, and checklist. Note a version date and a person responsible for maintenance.

Why this works

A colleague can now assess whether the prompt is suitable without trying all the variations. A name like 'final' says nothing about the actual quality.

Make it slightly more difficult

Find two cards that do almost the same thing. Decide which one to merge and which differences justify a separate card.

Extra practice, at your own pace. You do not need to submit anything. This does not count towards an automatic assessment.

Your workplace · learning by doing

Make something. See what changes.

Ende, your light pink practice buddy

The fictitious source is included. Only when selecting ‘Let Ende answer’ do you send your assignment to the AI service. Do not use private data.

Checking the AI connection… You can already view the written example.

Your result

No answer yet

Present your assignment to Ende.
Read the answer, change one thing, and compare.

You remain in control

What do you want to improve in your next attempt?

My own check

On this device only. AI answers remain only in this page memory. This is self-review, not a grade.

Why this approach works

A colleague can now assess whether the prompt is suitable without trying all the variations. A name like 'final' says nothing about the actual quality.

A step further

Find two cards that do almost the same thing. Decide which one to merge and which differences justify a separate card.

A prompt to get you started

Create a short description for this prompt so a colleague can find it: “[prompt]”. Provide a title, when to use it, and what to fill in.

Apply it

In a shared folder, there are multiple prompts with the same name. No one knows which version is still being used.

Multiple versions in a folder. Illustration for the question.

Which approach best fits this situation?

  • Only save the last modification date without explanation.
  • Add purpose, input fields, owner, and version to every prompt.
  • Remove the context so that the prompt becomes shorter.

With JavaScript, you can check your approach and receive immediate feedback.

Test your insight

Organising prompts together. Illustration for the question.

Why is a prompt library useful?

  • You don't have to reinvent good prompts every time.
  • AI will improve because of it.
  • It is mandatory.
Extra practice

Choose three prompts that you use frequently. Give them a title and note when they are useful.

Title: “Action list from meeting notes”. Use: after every meeting. Fill in: the raw notes.

A question? I'll help you on your way.

Drag Aldo to move it. Use the arrow keys on Aldo with the keyboard, Enter to save the position and Escape to cancel.