anything2explainer

Install

npx skillfoxx add skills/anything2explainer

npx skills add vincentwei1021/anything2explainer --skill anything2explainer -a claude-code -y

npx skills add vincentwei1021/anything2explainer --skill anything2explainer -a cursor -y

npx skills add vincentwei1021/anything2explainer --skill anything2explainer -a github-copilot -y

npx skills add vincentwei1021/anything2explainer --skill anything2explainer -a codex -y

npx skills add vincentwei1021/anything2explainer --skill anything2explainer -a gemini-cli -y

npx skills add vincentwei1021/anything2explainer --skill anything2explainer -a cline -y

npx skills add vincentwei1021/anything2explainer --skill anything2explainer -a roo -y

npx skills add vincentwei1021/anything2explainer --skill anything2explainer -a opencode -y

git clone https://github.com/Vincentwei1021/anything2explainer.git ~/.claude/skills/anything2explainer

This is third-party code. Review the repository files before installing.

What it does

The skill gives the agent the whole method for building an explainer video from a topic or an article. Every frame is drawn in code with Remotion in React and TypeScript, with no stock footage and no generative video model: a black canvas, white line graphics, a purple accent, large subtitles and a chapter progress bar at the bottom. It ships a compilable template, a library of visual primitives and lighting, tooling for voiceover, storyboard, rendering and quantitative quality checks, written style and motion specs, a multi-agent division-of-labour protocol and one finished reference film as the quality bar. Voiceover runs through TTS, by default edge-tts for Chinese and kokoro for English, with language and length set by the user. The agent pauses at four confirmation points: length and language, the final script, voiceover and the first thirty seconds.

Who it is for. For people who make science-communication and educational videos: content creators, teachers and marketers.

Example request

Make a 4-minute explainer video about how a CPU cache works, in English

Limitations

The license is PolyForm Noncommercial: use is limited to non-commercial purposes. Rendering needs Node.js, Remotion and Python, and building a clip takes noticeable time and compute. The default free voiceover is adequate but not studio grade; for better audio the author suggests a stronger TTS or a commercial API. Every number, term, year and name on screen must have a source in the research file; the skill does not fact-check for you.

How to disable. Remove the anything2explainer folder from the agent's skills directory, for example ~/.claude/skills/ or ~/.codex/skills/.

Security check

• Runs the Remotion build and Python scripts for rendering and quality checks
• Downloads packages and fonts, and for voiceover may reach a cloud TTS

README in short

The README describes anything2explainer as a Claude Code and Codex skill that turns a topic into an original explainer clip where every frame is code-drawn with Remotion. It is not a CLI but the whole method: a compilable template, a primitives library, tooling for voiceover, storyboard, rendering and quality checks, style and motion specs, a multi-agent protocol and one reference film about RAG and knowledge bases as the bar. The visual system is fixed: a black canvas, white lines, a purple accent, large subtitles and a chapter bar. README versions exist in English and Chinese. Licensed PolyForm Noncommercial.

SKILL.md

---
name: anything2explainer
description: Turn any topic into a narrated motion-graphics explainer video in Chinese or English, on a black canvas with a star-field or dot-field backdrop, TTS voiceover, subtitles and a chapter progress bar, every frame drawn in code with Remotion. Use when the user asks for an explainer, educational or science-communication video about a topic, or wants an article or document turned into a video.
---

# anything2explainer

Turn any technical or knowledge topic into an original explainer video. The visual system is fixed (black canvas, white line graphics with a purple accent, extra-bold type, 44px white subtitles with a black outline, a chapter progress bar at the bottom, a capsule HUD at the top); what changes is the content and the scale. Length is set by the user (confirmation point 1), and the narration, storyboard and shot code follow from it. Reference film: examples/rag/.

## When to use
- The user gives a topic and wants an explainer video, or gives an article/document to recast as a video.
- Not for: replicating an existing video frame by frame, a live presenter, or a clip built mostly from real footage.

## Hard rules
1. Original: every frame is drawn in code; optional B-roll must be royalty-free and logged in a manifest; no frames or clips from any existing video.
2. Facts have sources: every number, English term, year and name shown on screen must trace to a source URL in the research file.
3. One running example across the whole film, consistent across groups.

Related