teach

Generates structured HTML lessons and learning workspaces for teaching open-source projects.

2|Updated Jul 6, 2026
One-click install
npx skills add https://github.com/NAMEWTA/learning-open-code --skill teach-namewta
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: teach
Source: https://github.com/NAMEWTA/learning-open-code/tree/main/.agents/skills/teach
Command: npx skills add https://github.com/NAMEWTA/learning-open-code --skill teach-namewta

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Learning a large open-source project across multiple sessions is hard to organize: progress gets lost, lessons become bloated source-code dumps, and content goes stale when the upstream repository changes. This Skill provides a stateful teaching workspace that tracks your mission, learning records, resources, and versioned course snapshots. ## Core Features & Use Cases - Structured Teaching Workspace: Creates a two-level directory (project/topic) with MISSION.md, RESOURCES.md, SNAPSHOT.md, learning records, lessons, references, and shared assets, indexed per project. - Short-Lesson Contract: Enforces 15-minute HTML lessons with one learning objective, length limits, retrieval-practice exercises, and a completion audit via scripts. - Version Snapshot & Update Detection: Records the source repo's git commit and referenced files, then diffs against the current HEAD to flag which lessons need updates after the project changes. - Use Case: You want to learn the permission model of RuoYiVuePlus. The Skill groups its split submodules into one logical project, scaffolds the topic workspace, generates short HTML lessons with quizzes, snapshots the source commit, and later tells you exactly which lessons are stale after a git pull. ## Quick Start Ask the agent to teach you a specific topic from one of the open-source projects in this repository, for example: teach me the plugin system of claude-code.

Frequently Asked Questions about teach

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

FAQPage Schema
How do I start learning an open-source project with this Skill?▼

Tell the agent which project and topic you want to learn. It locates the matching entries in .gitmodules, groups split submodules into one logical project, scaffolds a topic workspace with init_topic.sh, and generates the first HTML lesson.

How are lessons kept short and focused?▼

Each lesson must teach one objective within 800-1500 Chinese characters, at most 4 sections, 3 source files, and 35-line code snippets, plus a retrieval exercise. The audit_topic.py script enforces these limits and rejects oversized lessons.

How does it detect when lessons are outdated after the source project updates?▼

Each topic stores a SNAPSHOT.md recording the source repo's git commit and referenced files. Running check_updates.py compares that commit with the current HEAD and diffs the referenced files to identify which lessons need revision.

What files track my learning progress across sessions?▼

Progress lives in the topic workspace: MISSION.md records your goal, learning-records/ stores numbered ADR-style insights, RESOURCES.md lists trusted sources, and NOTES.md captures teaching preferences. These files define your zone of proximal development for future lessons.

Can it handle projects split across multiple git submodules?▼

Yes. When several submodules share a common parent directory in .gitmodules, such as separate frontend and backend repos, the Skill treats the parent directory as one logical project and teaches them together under a single project root.