topic-teach

Generates structured Markdown courses teaching any knowledge topic with narrative lessons and reviews.

2|Updated Jun 8, 2026
One-click install
npx skills add https://github.com/HACK-WU/skills --skill topic-teach-hack-wu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: topic-teach
Source: https://github.com/HACK-WU/skills/tree/main/skills/topic-teach
Command: npx skills add https://github.com/HACK-WU/skills --skill topic-teach-hack-wu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Learning a new technical or non-technical topic with an AI often produces shallow, unstructured explanations that are hard to revisit. This Skill turns the AI into a structured tutor that plans a multi-stage curriculum, teaches in verified batches, and leaves behind a complete set of reusable Markdown learning materials. ## Core Features & Use Cases - Dual delivery modes: Full course mode (multi-stage outline, batched lessons of 3-5 knowledge points) for big topics, and quick-overview mode for single concepts. - Five-act narrative teaching: Every lesson follows scene introduction, cognitive conflict, layered revelation, hands-on verification, and system wrap-up, with analogies, Mermaid/SVG diagrams, and terminology anchoring. - Quality gates and review: Pre-writing official documentation checks, post-writing fact-checking, teaching-effectiveness rehearsal of the outline, and dual-perspective (pedagogy + learner) review before delivery. - Complete learning artifacts: Capstone project, troubleshooting handbook, scenario solution library, source-code walkthroughs, resume-from-breakpoint learning archives, and optional quizzes. - Use Case: Ask the AI to teach you Kubernetes from zero; it produces a staged course with lessons, SVG diagrams, hands-on exercises, a final project, and a troubleshooting manual saved as Markdown files you can keep and resume anytime. ## Quick Start Ask the AI to teach you a topic, for example: teach me Kubernetes from scratch using the topic-teach skill.

Frequently Asked Questions about topic-teach

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

FAQPage Schema
How do I use AI to systematically learn a new topic like Kubernetes?

Invoke the topic-teach skill with a request like 'teach me k8s'. It plans a multi-stage course outline, validates it with a teaching-effectiveness rehearsal, then generates lessons in batches of 3-5 knowledge points with reviews before each delivery.

What is the difference between course mode and quick overview mode?

Course mode handles complete subject systems with staged outlines, batched lessons, a capstone project, and troubleshooting manuals. Quick overview mode explains a single concept in one Markdown document with an extension outline, skipping the project and handbook phases.

Can I resume a course after closing the conversation?

Yes. All progress is stored in a learning archive file (00-学习档案.md) with a knowledge-point progress table and review records. Saying 'continue learning X' lets the skill locate the archive and resume from the first unfinished knowledge point.

Does topic-teach work for non-technical topics like personal finance?

Yes. It supports non-technical domains by replacing runnable commands with case walkthroughs, numeric comparisons, and decision exercises. High-risk domains like investing include mandatory risk disclaimers and avoid individualized advice.

How does the skill prevent outdated or fabricated technical content?

It enforces two gates: a pre-writing check against official documentation for every lesson, and a post-writing fact-check that verifies time-sensitive claims online with confidence annotations. Fabricated APIs, commands, or source code are prohibited.

When should I use module-teach instead of topic-teach?

Use module-teach when the teaching target is a code module in your current project, since it grounds explanations in actual code reading. Use topic-teach for general cross-project knowledge, which relies on built-in knowledge plus online verification.