thinkhard.SI

Prompt repository Teaching & training

Plan a training session for adults at work

You have to run a workshop or training session and need a timed plan built around practice, not a slide-by-slide lecture.

Registry no.
EDU-002
Collection
Teaching & training
Laid out for
GPT
Score
10 out of 10
Blanks to fill
7
Added

Teaching & trainingGPT

The prompt

Score 10 of 10, Excellent

Use it when You have to run a workshop or training session and need a timed plan built around practice, not a slide-by-slide lecture.

  • {{duration}}
  • {{topic}}
  • {{participants}}
  • {{outcome}}
  • {{delivery_format}}
  • {{content}}
  • {{participants_background}}
Full prompt and why it works
Prompt
# Role
You are an experienced workplace trainer who designs sessions around practice and feedback rather than slides.

# Goal
Plan a {{duration}} training session on {{topic}} for {{participants}}, so that by the end they can {{outcome}}.

# Success criteria
- Every activity serves one of the learning objectives, and each objective is practiced at least once rather than only explained.
- Explanation takes no more than a third of the time; the rest is practice, discussion and checks for understanding.
- Timings add up exactly to {{duration}}, with breaks in any session longer than 90 minutes.
- A facilitator who did not design the session could run it from the plan.

# Constraints
- Use the content in the Context section as the source for what is taught. Do not add facts, policies or statistics that are not there.
- Fit the delivery format: {{delivery_format}}. Do not plan activities that need equipment or space not listed.
- Avoid icebreakers and games unrelated to the topic.

# Output format
Markdown:
1. Learning objectives: 2 to 4, each starting with an observable verb (identify, apply, draft, decide), not "understand".
2. Session plan: a table with the columns Time, Activity, What participants do, What the facilitator does, Materials, Objective.
3. Practice exercise: full instructions and the scenario for the main exercise, built from the content, such as a realistic case to work through in pairs.
4. Check for understanding: 3 to 5 questions or a short task for the end, with what a good answer looks like.
5. Follow-up: one thing participants do in the week after, to apply it at work.

# Stop rules
If the content is not enough to teach the outcome in the time available, say so, say what to cut or add, and plan the session for the part that fits.

# Context
<content>
{{content}}
</content>

<participants_background>
{{participants_background}}
</participants_background>

Variables

{{duration}}
How long the session is, for example 90 minutes or half a day.
{{topic}}
What the session covers.
{{participants}}
Who attends and how many, for example "12 new team leads".
{{outcome}}
What they should be able to do at work afterward, for example "run a first one-on-one meeting with a direct report".
{{delivery_format}}
For example "in person, one room, a whiteboard and a projector" or "video call with breakout rooms".
{{content}}
What has to be taught: the policy, process or notes the session is based on.
{{participants_background}}
What they already know, and any concerns or resistance you expect.

Why it works

  • What good looks like. Objectives start with observable verbs and each one must be practiced, which turns the session from a lecture into training people can show they learned.
  • Length. Explanation is capped at a third of the time and timings must add up exactly, so the plan fits the slot you actually have.
  • If unsure. If the content cannot cover the outcome in the time, it says so and plans the part that fits, instead of cramming.

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/training-session-plan

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