| name | Setup New Exercise |
| description | Create a new exercise in one primary language and generate stubs for the others. |
Setup New Exercise Skill
Use this skill when the user asks to "create a new exercise" or "add a module".
1. Preparation
- Determine Topic: Choose a unique slug (e.g.,
exercise-bias-in-ocr).
- Select Primary Language: Ask the user which language they want to draft in (
en, de, or fr).
- Select Metadata:
- Category: Pick ONE.
- EN:
AI Literacy, Source Criticism, Ethics, Methods
- DE:
KI & Digitale Methoden, Quellen, Ethik, Methoden
- FR:
Littératie IA, Critique des sources, Éthique, Méthodes
- Tags: Pick 2-4.
- Resource Localization: Favor resources available in all three languages. If using language-specific resources, prepare to flag them or provide alternatives for the other language versions.
2. File Creation
Create three files: one primary draft and two stubs.
A. Primary Draft (Full Template)
File: [lang]/exercises/[slug].qmd
---
lang: [lang]
title: "Title"
description: "Description"
author:
- name: "Your Name"
affiliation: "Institution"
date: "today"
date-modified: "today"
categories: [Category]
tags: [Tag1, Tag2]
difficulty: [Difficulty]
time_estimate: "45 minutes"
draft: true
---
* Objective 1
* Objective 2
[Content]
[Instructions]
[Instructions]
* Question 1
::: {
:::
B. Language Stubs (Placeholders)
For the other two languages, create a file with minimal content to establish file parity.
File: [other_lang]/exercises/[slug].qmd
---
lang: [other_lang]
title: "TRANSLATION NEEDED: [Original Title]"
description: "Translation needed"
draft: true
---
:::{.callout-warning}
This exercise is currently only available in [Primary Language].
:::
3. Next Steps
- Inform the user that the primary draft and stubs are created.
- Suggest running the Translation skill to generate the missing versions.