← Back to Research Explained

Research Explainer

A Claude Code skill that turns AI research papers into interactive, beginner-friendly explainer pages.

Give it a paper URL, and it generates a self-contained HTML page with hover-defined terms, real-world analogies, architecture diagrams, HyperFrames motion graphics, and comprehension quizzes — all deployable to GitHub Pages.

Install

Run in your terminal
npx skills add https://github.com/terencecho/research-explained --skill skill

Use

Once installed, use it in Claude Code by providing a paper URL:

/research-explainer https://arxiv.org/pdf/some-paper.pdf

# or just say it naturally:
explain this paper: https://arxiv.org/pdf/some-paper.pdf

What It Produces

Aa
Every Term Defined
Hover tooltips on all AI/ML jargon, plus a searchable glossary panel.
💡
Real-World Analogies
Each concept grounded in an analogy a non-expert can relate to.
🎬
Motion Graphics
HyperFrames videos for concepts that are hard to picture from text alone.
🧩
Architecture Diagrams
Styled HTML card flows and pipeline visualizations — no ASCII art.
🧪
Comprehension Quizzes
Multiple-choice questions that test understanding of key concepts.
📱
Mobile Responsive
Works on phones and tablets. Sidebar becomes a slide-out drawer.

Examples

Pages built with this skill:

LoRA: Low-Rank Adaptation of Large Language Models
Microsoft • Parameter-efficient fine-tuning
Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
Mem0 • Persistent memory for AI agents
Fast KV Compaction via Attention Matching
MIT • KV cache compaction + multi-agent memory sharing
Avatar V: Scaling Video-Reference Avatar Generation
HeyGen Research • Video-conditioned avatar generation

Requirements

View on GitHub →