Skip to content
Agentmart
Submit a tool

learn

Use when someone wants to learn a subject, skill, language, or technology themselves over time, asks for a lesson, a study plan, practice questions, or a quiz…

Skill by jpenley

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/jpenley-learn

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when someone wants to learn a subject, skill, language, or technology themselves over time, asks for a lesson, a study plan, practice questions, or a quiz for themselves, or wants to pick up where their learning left off.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Node.js package

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/learn

What's inside

Skill name
learn
Description
Use when someone wants to learn a subject, skill, language, or technology themselves over time, asks for a lesson, a study plan, practice questions, or a quiz for themselves, or wants to pick up where their learning left off.
License
MIT
Instructions
41 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts

Files (6)

  • LEARNER.md
  • LESSONS.md
  • SKILL.md
  • openai.yaml · agents/
  • lesson.mjs · scripts/
  • review.mjs · scripts/

SKILL.md (first lines)

# Learn

Teaches the user something over several sessions: short interactive lessons that build on what they already know, and spaced practice that brings back what's fading. Progress lives in a learner folder, so the next session starts where this one stopped.

## 1. The learner folder

Look for `knowing.md` in the current folder or its parents. With none, ask where this subject's folder should live (a folder of its own, never inside a code repository) and create `knowing.md` there from [LEARNER.md](LEARNER.md). With one, read it: the goal, the level, the log, and what's due for practice (`node <skill>/scripts/review.mjs due`, with this skill's folder as `<skill>`).

Fill in **Goal**, **Level now**, and **Background to build on** from what the user has told you; ask one question only if the answer changes where to start.

## 2. The lesson

Write the next lesson as `lessons/<nn>-<topic>.md` following [LESSONS.md](LESSONS.md), then render it:

```
node <skill>/scripts/lesson.mjs lessons/<nn>-<topic>.md
```

Connect every new idea to the learner's background explicitly ("in Python, the garbage collector does this for you; in Rust, the owner does"). Show the mistake before the fix: a worked example that fails, the error it gives, and why. Keep it to one idea.

Give the user the `.html` file to open. Answers stay hidden until they check each one.

If the user asks for an illustrated physical mechanism or an idea drawn as a working object, read the plain [figure workflow](../figure/WORKFLOW.md) and follow it for that artifact. Its paths resolve from `../figure/`, not this skill's folder. Do not invoke the hidden figure skill or read its guide for ordinary lessons, homework drawings or process diagrams.

## 3. Practice

Add the lesson's two or three most important points to spaced practice:

```
node <skill>/scripts/review.mjs add --prompt "<question>" --answer "<answer>" --topic "<topic>"
```

At the start of a later session, run `review.mjs due` and ask those questions first, one at a time. An answer has three levels: recalled without notes; holding up under a follow-up (why it works, when it applies, how it fails, what else would do); tied to the learner's own work or goal. After a correct recall, ask one follow-up before recording; record `review.mjs record <id> right` only at the second level or above, otherwise `wrong`, and note the level reached in the log.

## 4. Log

Add a row to the log in `knowing.md`: what the lesson covered, what the learner got right and wrong, and what to revisit. Update the plan.

**Done when** the lesson's `.html` exists and renders without problems, its key points are in `practice.md`, and the log has today's row.

From jpenley/AITuning, MIT.

More from jpenley/AITuning

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 20, 2026 (9 days ago)
Last release
Oct 17, 2026 (12 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from jpenley

Cite this page

learn skill. Agent Mart. https://agentmart.dev/skills/jpenley-learn (last updated 2026-10-24)