teach

Creates stateful teaching workspaces with HTML lessons, learning records, and curated resources.

1|Updated Mar 2, 2025
One-click install
npx skills add https://github.com/marjorg/setup --skill teach-marjorg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: teach
Source: https://github.com/marjorg/setup/tree/main/home/.agents/skills/archive/0-matt/teach
Command: npx skills add https://github.com/marjorg/setup --skill teach-marjorg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Learning a new topic with an AI assistant is usually stateless: each session starts from zero, lessons are not grounded in your goals, and nothing tracks what you already know. This Skill turns a directory into a persistent teaching workspace that remembers your mission, your progress, and your trusted resources across sessions. ## Core Features & Use Cases - Stateful workspace files: Maintains MISSION.md (your learning goal), RESOURCES.md (curated high-trust sources), GLOSSARY.md (compressed terminology), and sequentially numbered learning records that capture demonstrated understanding and misconceptions. - HTML lessons and references: Produces short, self-contained HTML lessons in ./lessons/ plus printable reference documents in ./reference/, built from reusable components in ./assets/ for a consistent course look. - Zone of proximal development targeting: Reads your learning records to pick the next lesson that challenges you just enough, using retrieval practice, spacing, and interleaving for long-term retention. - Use Case: You want to learn strength training over several weeks. The Skill interviews you to write a mission, curates evidence-based resources, generates an interactive HTML lesson on progressive overload with a quiz, and records what you mastered so the next session builds on it. ## Quick Start Ask the agent to teach you a topic, for example: teach me the basics of strength training in this workspace.

Frequently Asked Questions about teach

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I use an AI agent to learn a new topic over multiple sessions?

Invoke the teach skill in a dedicated directory and state what you want to learn. The agent creates a teaching workspace with a mission file, curated resources, and sequential HTML lessons, recording your progress in learning records so each session builds on the last.

How does the teach skill decide what to teach next?

It reads your learning records and MISSION.md to estimate your zone of proximal development, then teaches the most relevant concept that challenges you just enough. If your mission is unclear, it interviews you first before writing any lesson.

What files does the teaching workspace contain?

The workspace holds MISSION.md for your goal, RESOURCES.md for trusted sources, GLOSSARY.md for terminology, NOTES.md for preferences, a learning-records directory of numbered markdown insights, a lessons directory of HTML lessons, a reference directory of printable cheat sheets, and an assets directory of reusable components.

Can I use this skill for one-off questions instead of long-term learning?

It is designed for stateful, multi-session learning and sets disable-model-invocation, so it only runs when explicitly invoked. For quick factual questions, a normal conversation is more appropriate than creating a full teaching workspace.

Why does the skill prefer external resources over the model's own knowledge?

The skill instructs the agent to never trust parametric knowledge and to ground lessons in high-trust sources tracked in RESOURCES.md. Lessons include citations to these sources, which increases accuracy and lets you verify claims independently.