thinkhard.SI

Prompt repository Teaching & training

Write a branching scenario for workplace training

You want learners to practice a decision they face at work, with realistic choices and consequences instead of right-or-wrong quiz feedback.

Registry no.
EDU-016
Collection
Teaching & training
Laid out for
Claude
Score
10 out of 10
Blanks to fill
6
Added

Teaching & trainingClaude

The prompt

Score 10 of 10, Excellent

Use it when You want learners to practice a decision they face at work, with realistic choices and consequences instead of right-or-wrong quiz feedback.

  • {{source_material}}
  • {{learners}}
  • {{objective}}
  • {{delivery}}
  • {{common_mistakes}}
  • {{minutes}}
Full prompt and why it works
System prompt
You are an instructional designer who writes realistic workplace scenarios. Your scenarios put learners in a situation they recognize, offer choices that all seem reasonable, and show consequences instead of lecturing.
User message
<source>
{{source_material}}
</source>

<design>
Learners and their job: {{learners}}
Skill or decision to practice: {{objective}}
Where it will be built (tool or format): {{delivery}}
Mistakes people make in real life, from managers or experts: {{common_mistakes}}
</design>

Write a branching scenario that practices the decision above, so that learners can try it safely before they face it at work. It is a draft for an instructional designer to build in {{delivery}} after a subject matter expert has checked it.

Structure:
1. Setup: under 100 words, in second person, with a named character and a realistic moment from the learner's job.
2. Three decision points. At each, offer three choices: one best, one partly right, and one common mistake from the list above. All three should look reasonable to someone who has not mastered the skill.
3. For each choice, a consequence shown through what happens next (what the customer says, what the colleague does, what the numbers show), under 60 words, followed by one sentence of coaching that explains why.
4. Endings: three, from best to poor, each under 80 words, with what the learner should take away.
5. A branch map: a table with the columns Node, Situation, Choice, Leads to.

Keep the whole scenario playable in about {{minutes}} minutes. Paths may rejoin after a weak choice, so the scenario stays manageable to build.

<example>
Feedback that only labels: "Incorrect. You should acknowledge the customer's concern." Feedback that shows a consequence: "The customer sighs. 'So you're not going to look into it?' Her tone has changed, and the call is now harder to recover."
</example>

Take procedures, rules and facts from the source material only. Where the story needs a realistic detail the source does not give, mark it [SME to confirm]. Avoid stereotypes in names, roles and dialogue, and use a varied mix of names. If the objective covers more than one decision, say so and pick the one most worth practicing.

Variables

{{source_material}}
The policy, procedure or expert notes the right answers come from.
{{learners}}
Who the learners are and their job, for example "new call center agents in their first month".
{{objective}}
The decision to practice, for example "handle a customer who demands a refund outside policy".
{{delivery}}
The tool or format, for example Articulate Storyline, Rise, or a printed role-play.
{{common_mistakes}}
What people really get wrong in this situation, from managers, quality reviews or experts.
{{minutes}}
How long the scenario should take, for example 10.

Why it works

  • An example. A pair of feedback examples shows the difference between labeling an answer wrong and showing what happens next, the core of a good scenario.
  • What good looks like. Every decision point has a best, a partial and a common-mistake choice that all look reasonable, so learners have to think rather than spot the obvious answer.
  • If unsure. Details the source does not supply are marked for the subject matter expert, because invented procedure in training gets learned as fact.

How to use it

  1. Press Fill in variables and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/branching-scenario-for-training

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "branching-scenario-for-training"}, or just ask for it by name.