Prompt library
Build a learning plan for a new skill
Deconstruction plus feedback-source-per-sub-skill applies the two strongest findings in skill acquisition: decompose and shorten the error-correction loop. Scheduling a real output in week one and pre-banning the classic time sinks protects the scarce resource, which is motivation, not information.
Last reviewed July 17, 2026
The prompt
Build a learning plan.
The skill: {{skill}}
Why I want it and the level that would satisfy me: {{target}}
Time budget: {{time}}
Produce:
1. Deconstruct: the 4-6 sub-skills this actually decomposes into, ordered by which unlocks the others. Name the one sub-skill that is the usual bottleneck for adults learning this.
2. The practice spine: for each sub-skill, the practice activity (not the reading; the doing) and the feedback source that tells me I am wrong quickly (an app that corrects, a checker, a person, recording myself). Learning rate is feedback rate.
3. The 20-hour arc: what weeks 1-2 look like session by session, designed so the first real, satisfying output happens inside week one (motivation runs on evidence).
4. The plateau plan: where progress typically stalls for this skill and the drill that breaks the specific stall.
5. What to ignore: the topics beginners waste time on for this skill (gear, theory rabbit holes, tool debates), banned until a named milestone.
6. The proof: how I will demonstrate the target level to myself (a concrete performance, not a feeling).
Optimize for time-to-first-real-output, then for feedback density.Run in idaptOpens a new chat with the prompt prefilled. Nothing sends until you press send.
Fill in the variables
| Variable | What it is | Example |
|---|---|---|
| {{skill}} | What you want to learn | conversational Spanish |
| {{target}} | Why, and satisfying level | family trips twice a year; hold a 10-minute everyday conversation |
| {{time}} | Real budget | 30 minutes daily on the commute plus one hour Sunday |