learn-topic

Builds durable multi-session learning workspaces with lessons, reference docs, and progress logs.

Updated Jul 3, 2026
One-click install
npx skills add https://github.com/mnazaal/dotfiles --skill learn-topic-mnazaal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: learn-topic
Source: https://github.com/mnazaal/dotfiles/tree/main/.agents/skills/learn-topic
Command: npx skills add https://github.com/mnazaal/dotfiles --skill learn-topic-mnazaal

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? One-shot explanations of a new topic are forgotten quickly and leave no record of what you already understand. This Skill turns a learning goal into a persistent workspace with numbered lessons, a growing cheat-sheet reference, and a log of insights and corrected misconceptions that carries across sessions. ## Core Features & Use Cases - Structured Lesson Generation: Writes self-contained, single-sitting HTML lessons numbered from 0001, each grounded in a stated mission and citing an external source. - Durable Reference Docs: Maintains a sectioned CONTEXT.md cheat-sheet for five-second lookup and a prepend-only LOG.md of dated insights that calibrates the next lesson's difficulty. - Zone of Proximal Development: Reads the learning log before each lesson to teach the most relevant next thing that is challenging but not overwhelming. - Use Case: You want to ramp up on distributed systems over several weeks. The Skill creates a topic workspace, interviews you on why, then produces a sequence of cited HTML lessons while tracking what you have already mastered. ## Quick Start Ask the assistant to teach you a topic, for example: "Help me learn the basics of Kubernetes networking over the next few sessions."

Frequently Asked Questions about learn-topic

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

FAQPage Schema
How do I learn a new topic across multiple sessions with an AI assistant?▼

State your learning goal and the Skill creates a topic workspace with a MISSION.md, then writes one self-contained HTML lesson per session. A LOG.md records insights and corrections so each new lesson builds on what you already understand.

How does the Skill decide what to teach next?▼

It reads LOG.md before each lesson to target the zone of proximal development, choosing the most relevant next concept that is challenging but not overwhelming. The mission in MISSION.md grounds every lesson's scope.

Does the Skill teach from the model's own knowledge?▼

No. Each lesson requires finding and citing a high-quality external source before writing, so content is never based on parametric knowledge alone. Academic topics can route citations through a research-protocol skill.

What files does the learning workspace contain?▼

The workspace holds MISSION.md for the learning goal, CONTEXT.md as a sectioned cheat-sheet reference, LOG.md as a prepend-only dated insight log, numbered HTML lessons under lessons/, and shared assets/ for reusable lesson components.

When should I use this instead of asking for a direct explanation?▼

Use it when you want durable, progressive understanding of a subject over time rather than a one-off answer. For understanding a specific project or codebase, use a codebase-understanding skill instead, since this Skill teaches topics from external sources.