Deep dive · module 21 of 100
Build a prompt library
Save good prompts and make them reusable for yourself and your team.
After this lesson
You create a simple structure to store, name, and share prompts.
In view · How it connects
A prompt as a reusable recipe
Save a prompt that you have already tested on a small task.
Replace varying data with clear placeholders.
Note goal, version, owner and what you need to check.
Part 1 of 3 · Working example
Your preference for less movement is on. Choose each part yourself.
Why save them?
Good prompts disappear in chat history. A library makes them repeatable and improvable. Do not just save a successful answer, but also the input intended for the prompt. A prompt for a short customer email does not automatically work for a technical report. Note which human control belongs to the application.
Simple structure
Give each prompt a name, short description, purpose, and the prompt itself. Add an example result if necessary. In addition to name and date, add the required input, desired output, owner, and control criteria. Use placeholders for private data. Include a fictitious test case so that someone else understands how the prompt was tested.
Sharing with the team
Put the library in a shared document or Notion. Agree on who can add what and how to update outdated prompts. Limit editing rights where necessary and agree on a review moment. Mark an outdated variant as withdrawn. Do not share chat history containing customer data just because the prompt itself is useful.
New practice example
From explanation to a usable result.
Do not just save the prompt, but also the goal, the permitted input, and an example of good work.
Fictional practice situationFictional: a team has a good prompt for a weekly update. A new colleague does not know which notes are allowed in it.
View a possible solution
This is how it can be done
Create a sheet with task, input fields, fictional input example, expected format, checklist, owner, and version date. Indicate which data should not be included in the input.
Why this works
The sheet makes it clear when the prompt is suitable and how to assess a change.
Make it slightly more difficult
Create a second version for a different audience. Which components remain common and which must be managed separately?
Extra practice, at your own pace. You do not need to submit anything. This does not count towards an automatic assessment.
Watch and understand · with narration
From vague to useful
With a computer-generated voice, on-screen text and captions in your chosen language. Start and pause whenever you like. Open the video separately.
Read the text version of the video
- A vague request: write an email. The goal, reader and facts are missing.
- Give context: an invitation to a fictional online workshop, on Saturday from ten to eleven.
- Ask for no more than ninety words, plain language and one next step. Leave missing information open.
- Check the time, invented claims and the next step. Then ask for one specific improvement.
Your workplace · learning by doing
Make something. See what changes.
Checking the AI connection… You can already view the written example.
Present your assignment to Ende.
Read the answer, change one thing, and compare.
Previously held
New version
You remain in control
What do you want to improve in your next attempt?
On this device only. AI answers remain only in this page memory. This is self-review, not a grade.
Why this approach works
The sheet makes it clear when the prompt is suitable and how to assess a change.
A step further
Create a second version for a different audience. Which components remain common and which must be managed separately?
A prompt to get you started
Help me set up a prompt library for [team or task]. Create a table with columns: name, purpose, prompt, last update. Provide three examples. Use only the provided fictitious facts. Make missing information visible and indicate what a human must check before use.
Apply it
Your colleague finds an old prompt "write a customer email". No one knows for which task it was approved or what data it expects.
Which library entry makes the prompt usable for the team?
- Only a title and five stars.
- Purpose, required input, prompt, fictitious test case, control criteria, owner, and version date.
- A full chat with real customer data.
With JavaScript, you can check your approach and receive immediate feedback.
Test your insight
What is the main advantage of a prompt library?
- AI automatically gets better.
- Good prompts become repeatable and shareable.
- You never have to check again.
Extra practice
Write three prompts that you use regularly or want to use. Give each prompt a short name and describe what you use it for. Work with a fictitious situation and name one concrete control point before you would use the result.
Name: Customer email follow-up. Purpose: friendly follow-up after a conversation. Prompt: Write a short follow-up email... I compare the outcome with the confirmed input, leave unknown data blank, and adjust the result before use.
