Copy-and-paste · verified July 27, 2026
Learning and teaching prompts for Claude Fable 5
Six scaffolds for explaining, structuring a curriculum and generating practice — aimed at finding the gap, not filling the silence.
What actually helps here
Say what you already know and where it stops. An explanation pitched at the wrong level is the most common waste here, and a model cannot guess your level from the question alone. Naming the adjacent thing you do understand gives it an anchor to build from, and asking it to check your stated understanding first catches the misconception before the explanation lands on top of it.
The prompts
Fill in the bracketed slots, paste into any Claude app or the API. Written for Fable 5's goal-first style — they state the outcome and the constraints rather than scripting the steps.
Explain to a beginner
Explain [concept] to someone new to the field. Use one everyday analogy, one concrete example, and end with the single most common misconception.
Study plan
Build a [4]-week study plan for learning [skill] at [2] hours per week. Each week: a goal, the type of resource to use, and one self-test. Assume no prior background.
Quiz generation
Create a 10-question quiz on [topic]: 6 multiple choice, 3 short answer, 1 applied scenario. Include an answer key with one-line explanations.
Socratic tutor
Act as a Socratic tutor on [topic]. Ask me one question at a time, adapting to my answers. Never explain directly until I ask or fail twice.
Analogy finder
Give three analogies for [concept] drawn from [cooking / sports / city planning], and note where each analogy breaks down.
Misconception check
Here is my understanding of [topic]: [paste]. Identify what's right, what's subtly wrong, and what's missing — in that order.
Getting more out of these
Two habits matter more than the wording. Replace every bracketed slot before you send — Fable 5 follows instructions literally, and an unfilled [topic] becomes part of the task. And resist adding "think step by step" or "show your reasoning": on Fable 5 that phrasing can trip the reasoning_extraction classifier and hand your request to Opus 4.8 instead.
Need something these do not cover?
The generator builds a full system prompt around your own task, autonomy level and output style, and hands back the API call to go with it.