Learning Path Builder
Self-directed learning fails not from lack of motivation but from lack of structure. The learner doesn't know what order to learn things in, doesn't know what 'good enough' looks like at each stage, picks resources that are wrong for their level, and has no way to measure progress. The Learning Path Builder fixes all four — and the first step is achievable today, in under 30 minutes.
What this skill does
The first move is an interview, not a curriculum. What specifically do you want to learn — not "Python" but "Python for data analysis so I can automate my weekly reports". What do you already know — adjacent knowledge changes the path entirely, because a JavaScript developer learning Python skips most of Phase 1 and lands in a 1-day syntax bridge instead. How much time can you commit honestly — "I'll study 4 hours a day" is a fantasy for most working adults and a path designed for it will fail. The skill plans for 30 minutes a day, 1 hour, or 2+, and the curriculum compresses or expands accordingly.
The architecture is five phases. Phase 0 is the quick win — something achievable in 30 minutes that ends with "I just did the thing" rather than "I just read about the thing", because motivation is highest at the start and the path needs to convert that motivation into a concrete experience before it fades. Phase 1 builds the minimum viable foundation, not an exhaustive survey — moving on when you can define each core concept in your own words, recognise it in an example, and answer a basic "when would you use this?" question. Phase 2 is structured practice with guided projects of increasing complexity. Phase 3 is an independent project with no tutorial. Phase 4 is the ongoing deepening that turns competence into fluency.
The skill is opinionated about resources: one primary resource per concept, not five. Choice paralysis kills momentum, and the user asked for a path, not a bibliography. Each resource ships with a "skip if..." note so you don't waste time on material you don't need.
Two things make the path actually finishable. Testable milestones — "build a responsive landing page that works on mobile without using a framework" beats "understand CSS". You know unambiguously whether you've hit it. And named plateaus — every learning path has a frustrating period around weeks 3-4 where progress feels invisible. The skill names it before you hit it, because the dip is a sign of progress, not failure, and the people who quit at the dip are the ones who didn't know it was coming.
When your timeline doesn't match your goal, the skill says so directly and offers two things: an honest estimate for the actual goal, and a realistic plan for the timeline that achieves the most valuable subset. Pretending unrealistic timelines are fine is how learners end up failing and blaming themselves instead of the plan.
When this triggers
- ·You want to learn a new skill and need a real plan, not a 47-tab Pinterest of resources
- ·You're switching careers and need a curriculum scoped to a specific job, not a whole field
- ·You've started learning something three times and quit because you couldn't tell if you were making progress
- ·You have an external deadline and need a compressed path that drops the nice-to-haves
- ·You want testable milestones so you know when to move to the next phase, not just feel ready
Example
Trigger
User: 'I want to learn Python for data analysis. I'm a marketer with no coding background. 1 hour/day.'
Output
## Learning Path: Python for Data Analysis (marketer, 1 hr/day) Time to competence: ~14 weeks Prerequisites: None — this path starts from zero. ## Phase 0: Quick Win (today, 30 min) Open Google Colab. Paste a 4-line script. Load a CSV of your last 6 months of marketing data. Print the column names. Done. ## Phase 1: Foundation (3 weeks) Core concepts: variables, lists/dicts, loops, functions, pandas DataFrames. One primary resource: [specific course], 12 hours. Milestone: load a CSV, filter for one condition, output a summary. Move on when you can do this without referring to the tutorial. ## Phase 2: Structured Practice (5 weeks) Projects: (1) analyse your campaign data and find which channel drove the most conversions; (2) build a script that produces your weekly report automatically. ## Phase 3: Independent Application (4 weeks) Build the thing you actually wanted Python for. No tutorial. ## Phase 4: Deepening (ongoing) [Resources, community, habits.] Motivation dip: around week 4. You'll know enough to see how much you don't know. Push through one more project before quitting — the dip is a sign of progress.
Get this skill + 7 more
Get the full Learning & Growth pillar (8 skills) or the complete library.
Get the full stack — $299What you get
- 231-line SKILL.md, ready to drop into ~/.claude/skills/
- Tested through 3 Karpathy-loop iterations (versions v1.0.0 → v1.3.0)
- Triggers automatically when relevant — no command to remember
- Lifetime updates as the skill is refined further
More from Learning & Growth
Takes a topic — any topic — and map the landscape of ideas within it: what the core concepts are, how they relate to each other, which ones are foundational (everything depends on them) and which are…
When someone presents a position, argues the other side — not as a lazy contrarian, but as the most informed, thoughtful version of the opposition
Takes any concept — technical, scientific, business, philosophical — and explain it at three distinct levels of complexity
Takes any learning material — notes, articles, textbook chapters, lectures — and create flashcards optimised for long-term retention
Doesn't quiz to confirm what someone knows — it quizzes to find what they *don't* know and force them to confront it
When someone asks "How good am I at X?", most people lie — friends encourage, colleagues hedge, and AI assistants default to flattery
Browse the full library
297 skills across 31 categories. One purchase, lifetime updates.
See all bundles