learning-lab-authoring

Enforces authoring standards for MDX lessons, formulas, and illustrations in Learning Lab content.

4|Updated Feb 17, 2026
One-click install
npx skills add https://github.com/duongtruongbinh/TorchViz-3D --skill learning-lab-authoring-duongtruongbinh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learning-lab-authoring
Source: https://github.com/duongtruongbinh/TorchViz-3D/tree/main/.agents/skills/learning-lab-authoring
Command: npx skills add https://github.com/duongtruongbinh/TorchViz-3D --skill learning-lab-authoring-duongtruongbinh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writing consistent, pedagogically sound technical lessons across many AI domains is hard: terminology drifts, formulas lack concrete numbers, and prose replaces visuals. This Skill standardizes how Learning Lab lessons, blog posts, and deep-dives are authored in TorchViz-3D. ## Core Features & Use Cases - Pedagogical Frameworks: Enforces domain-specific lesson structures (architecture deep-dives, math foundations, MLOps workflows) with mandated intuition-to-formula progression. - Terminology & Typography Rules: Governs bold usage, English technical term retention, callout restrictions, and precise parameter-scale phrasing. - Math & Visual Standards: Requires dimension-annotated BlockMath formulas, concrete numerical substitutions, and visual anchors (Flowchart, ConceptHierarchy, MetricBars) replacing prose. - Use Case: When drafting a new MDX lesson on LoRA for the Deep Learning domain, apply this Skill to structure the narrative, format SVD formulas with tensor shapes, select the right visual components, and keep lessonMetadata synchronized with the table of contents. ## Quick Start Use the learning-lab-authoring skill to draft a new Learning Lab lesson on attention mechanisms following the universal authoring standards.

Frequently Asked Questions about learning-lab-authoring

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

FAQPage Schema
How do I structure a Learning Lab lesson for a new AI topic?

Follow the domain-specific framework: architecture topics use Context, Representation, then Mathematical Solution; math foundations open with geometric intuition before algebra; MLOps topics start from production needs. Every section needs introductory prose before subheadings.

How should mathematical formulas be formatted in MDX lessons?

Use BlockMath components on isolated display lines with underbrace dimension annotations for tensor shapes. Always substitute concrete values from real models like LLaMA 7B and quantify parameter ratios, choosing bottom-up or top-down pedagogical flow.

When should I use Flowchart versus ConceptHierarchy components?

Use Flowchart for sequential step-by-step processes like pipelines and training loops. Use ConceptHierarchy for taxonomies and multi-level classifications with numbered leaf nodes. Never place two ConceptHierarchy trees in one section.

Can I use LessonNote callouts in lesson content?

Only when the user explicitly requests them. By default, present all notes, warnings, and takeaways as regular markdown prose. If a callout is requested, keep its text in regular weight with no bolding.

What catalog synchronization do MDX lessons require?

The lessonMetadata title must match the table-of-contents locale title character-by-character, and headings must list the exact section headings in the content. Published content must pass the npm run verify check.