anything2explainer Checked

Name: anything2explainer
Author: Vincentwei1021

给一个主题，产出一条黑底 MG 风格（幕底可选星点或点阵波）、有配音字幕章节进度条的科普讲解视频（中文或英文；Remotion 代码动画；时长由用户定，常用 3–5 分钟）。内含可编译模板、图元…

🎬 Video & Motion skills By Vincentwei1021 Version v1.0.0 Updated 2026-09-17

9.4Overall rating

Installation

🤖 Install via AI

Copy the prompt below and send it to your AI assistant (e.g. Claude Code) — it will follow the instructions and install automatically.

Install the "anything2explainer" skill by following the instructions at https://skill123.me/install/anything2explainer.

⌨️ Command line install

Run in your terminal — downloads and installs to ~/.claude/skills/.

curl -fsSL https://skill123.me/install/anything2explainer.sh | bash

📦 Download ZIP

Download the zip and extract it into your skills directory (e.g. ~/.claude/skills/), then restart your session.

⬇ Download v1.0.0 · 14.9 MB

About this skill

Overview

给一个主题，产出一条黑底 MG 风格（幕底可选星点或点阵波）、有配音字幕章节进度条的科普讲解视频（中文或英文；Remotion 代码动画；时长由用户定，常用 3–5 分钟）。内含可编译模板、图元库、配音/分镜/渲染工具、风格与动效规范、多 agent 分工协议与 QC 判据，以及一条完整样片（《RAG 与知识库》）作为质量标尺。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.

Source

Repo: https://github.com/Vincentwei1021/anything2explainer
Path: (repo root)

Score breakdown

Trigger

9.0

Bilingual What/When/When-Not description with explicit non-goals (video-replica, live-action); strong topic-video trigger vocabulary though fewer enumerated user phrasings than the best descriptions.

Structure

9.3

88-line SKILL.md offloading to 10 reference docs and a compilable Remotion template; clean scripts/reference/template/examples split; repo-level README/CITATION artifacts cost the no-redundancy point.

Workflow

9.6

Exemplary execution design: four mandatory stop-and-wait confirmation points (duration/language, script sign-off, TTS preference, 30s preview), parallel agent builds with per-group motion_check thresholds, selfcheck/static audits and final-frame re-measurement as the verification loop, documented lessons and retry handling for flaky edge-tts.

Content

9.0

Extremely precise quantified writing (frame budgets, px thresholds, words-per-second rates, disk estimates); density is very high, which costs some scannability.

Engineering

9.0

Valid frontmatter, shebangs present, all referenced paths exist, deps (Remotion 4.0.507, ffmpeg, TTS engines) pinned and documented; heavy environment assumptions (node, ffmpeg, espeak-ng, ~2GB disk) though all disclosed.

Security

10.0

Capability manifest matches declared scope: cloud TTS egress (narration text to Microsoft endpoint) is prominently disclosed and user-gated with local-only alternatives; new_project.sh validates slug against shell-injection; research docs are explicitly treated as data-not-instructions (anti prompt-injection rule); no credentials, no eval/dynamic code, no persistence.

Best-in-batch video-generation pipeline (source: github.com/Vincentwei1021/anything2explainer; 15MB site zip download throttled/unstable, assessed from repo). Bilingual Remotion explainer-video factory: 8-stage pipeline, 4 mandatory user confirmation gates, multi-agent build/QC orchestration with quantified motion/composition oracles, full worked example film. Capability manifest (6.0): reads only project files and user-approved narration; side effects confined to user-chosen project dir (rsync template, npm install ~2GB, renders/frames); network destinations = npm registry, Microsoft edge-tts cloud (sends narration text, explicitly disclosed at confirmation point 3 with local kokoro/piper alternatives and user-supplied-wav bypass), and agent research fetches of user-topic URLs; no credentials; no persistence. Capabilities match declared scope.