visually-explain

Generate a self-contained interactive HTML document that visually explains a specific codebase mechanism.

3|Updated Apr 14, 2026
One-click install
npx skills add https://github.com/nvie/skills --skill visually-explain-nvie
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: visually-explain
Source: https://github.com/nvie/skills/tree/main/skills/visually-explain
Command: npx skills add https://github.com/nvie/skills --skill visually-explain-nvie

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Static prose explanations of complex code mechanisms often fail to build accurate mental models, especially for visual learners. This Skill turns a concrete system, mechanism, or design from your codebase into a single interactive HTML document with toggles, steppers, diffs, and side-by-side comparisons that walk a real worked example through its states. ## Core Features & Use Cases - Interactive Explanations: Produces one self-contained HTML file with inline CSS and vanilla JS — no build step, no network calls, opens offline by double-clicking. - Verified Correctness: Requires hand-tracing a concrete worked example against the actual source code before building, baking verified states into the artifact rather than trusting live computation. - Matched Interactions: Chooses the right interaction pattern for the concept — compare toggles for divergences, steppers for protocols and lifecycles, side-by-side views for design tradeoffs, and live playgrounds for transformations. - Use Case: You are trying to understand how client-side rebasing diverges from server-side rebasing in a collaborative editor. Invoke the Skill to get an HTML page that threads a real "Hello" edit through both approaches step by step, with color-coded states and a toggle between current and intended behavior. ## Quick Start Ask the AI to run /visually-explain on a specific mechanism in the codebase, such as "visually explain how the rebasing logic in livetext works".

Frequently Asked Questions about visually-explain

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

FAQPage Schema
How do I create an interactive HTML explainer for a code mechanism?▼

Invoke /visually-explain with a specific mechanism you can point at in code or a spec. The Skill reads the source, hand-traces a concrete worked example, and writes one self-contained HTML file with steppers, toggles, or side-by-side views, then opens it in your browser.

What kinds of concepts can this interactive explainer visualize?▼

It explains particular systems grounded in actual code, design docs, or specs — protocols, algorithms, request lifecycles, and design tradeoffs. It does not produce general-concept tutorials, since without a concrete source there is nothing to verify the visual against.

Does the generated HTML file need a build step or internet connection?▼

No. The output is a single .html file with all CSS and JS inlined, using vanilla JavaScript with no frameworks or network calls. It opens by double-clicking and works fully offline, with at most one CDN web font that degrades gracefully.

How does the explainer avoid showing incorrect information?▼

Every state, transition, and label must be verified against the source code before inclusion, and states are baked in as hand-checked data rather than computed live. Genuine uncertainties are visibly marked as assumptions in the artifact instead of being presented as fact.

When should I not use a visual explainer for a concept?▼

Skip it for general-concept questions with no concrete code or spec to verify against — answer those in prose instead. It is also never auto-triggered; it only runs on an explicit /visually-explain request since it writes a file and opens a browser tab.