What does this skill do, and when should you use it?
anything2explainer is a Claude Code / Codex skill that turns any topic or article into a 1280×720 H.264 Remotion explainer video. It runs a full pipeline — sourced research, narration, TTS voiceover with word-boundary timelines, storyboarding, parallel shot building by multiple agents, rendering and quantitative QC — in roughly 1–3 hours of wall clock. Every frame is drawn in code with no generative video model and no footage lifted from existing videos, and every number on screen must trace to a source URL. The run stops at exactly four checkpoints (length/language, narration sign-off, voiceover engine, first 30 seconds) instead of ploughing through.
- Scaffolds a compilable Remotion 4 project from the template (npm install + tsc)
- Dispatches a researcher agent to produce a sourced research doc with a numbers-and-analogies list
- Writes narration, then runs tts_build.py for voiceover wav, word-aligned timeline.ts and subtitle tables
- Dispatches 4–14 parallel build agents (per the length table) writing one Remotion component per shot
- Renders the film (npx tsc --noEmit + render.sh) and runs frame_metrics / motion_check / selfcheck quantitative QC
- Runs one QC agent per chapter, fix agents per group, re-renders v2/v3, then ships delivery notes
- Content creators who want a few-minute educational video explaining a technical concept such as vector databases
- Authors with a technical article or document they want turned into a narrated, subtitled video
- Education/science-communication teams that need a consistent visual style with on-screen figures traceable to sources
- Channels that want reproducible Chinese or English explainer videos and accept the fixed black-canvas MG style
- Replicating an existing video, talking-head presenter footage, or mostly live-action films — explicitly out of scope
- Commercial use of the toolkit is restricted: PolyForm Noncommercial license, commercial use needs prior author authorization
- Landscape 1280×720 only; no 9:16 vertical support, and Windows is untested
How do you install this skill?
- This is a static source review only; nothing was executed, so render quality and efficiency claims are not independently verified.
- The default Chinese voice (edge-tts) calls a Microsoft cloud endpoint whose mainland-China reachability is undisclosed; the English kokoro model downloads from overseas sources — verify your network environment.
- Each film consumes ~2–3GB of disk and parallel builds are demanding (≥5GB free, tmux pane caps); no automatic rollback/cleanup path is documented.
- License is PolyForm Noncommercial: free for noncommercial use, commercial use requires prior author authorization; the repo's NOASSERTION metadata conflicts with the LICENSE text.
- Narration is frozen once voiced (hard-coded frame numbers); finalize wording at checkpoint 2.
- Verified only on macOS (plus Raspberry Pi ARM); Windows untested; only Claude Code/Codex have been actually run.
- Shell / CLI
- Network access
- Local filesystem
Node.js 18+ffmpegPython 3 (edge-tts, numpy, pillow, scipy)kokoro + soundfile + espeak-ng (English narration only)~2-3 GB disk per film
Clone the repo and symlink it for your host:
Claude Code
git clone https://github.com/Vincentwei1021/anything2explainer.git
ln -s "$PWD/anything2explainer" ~/.claude/skills/anything2explainerCodex
git clone https://github.com/Vincentwei1021/anything2explainer.git
ln -s "$PWD/anything2explainer" ~/.codex/skills/anything2explainerDependencies: Node ≥18, ffmpeg, and a Python venv with edge-tts==7.2.8 / numpy / pillow / scipy (English narration also needs kokoro, soundfile, espeak-ng).
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Make me an explainer video about vector databases
- 讲一下向量数据库，做成一条讲解视频
- Turn this RAG article into a 4-minute English explainer video
- Produce a 5-minute motion-graphics explainer on knowledge bases in English
Trigger it with one line in Claude Code or Codex (e.g. "Make me an explainer video about vector databases") and it walks the 9 stages in SKILL.md: scaffold → research → narration & timeline → storyboard → overlays & primitives → 30-second pilot → parallel build → render → QC & fixes. It stops at four checkpoints for your sign-off: (1) length and language, with no defaults; (2) narration sign-off — once locked, frame numbers are hard-coded and any reword re-times the whole film; (3) voiceover engine (defaults: edge-tts Yunxi for Chinese, kokoro am_liam for English, or bring your own finished wav); (4) the first 30 seconds of render. You can also drive the template manually: run new_project.sh, then tts_build.py → render_storyboard.py → preview.sh 30 → render.sh in order.
What are this skill's strengths and limitations?
- 100% code-drawn, reproducible frames (same frame number renders identically) with no generative video model
- Strong fact discipline: every on-screen number or term must trace to a source URL in the research doc
- Three quantitative QC scripts (composition/light, motion density, static self-check) plus a full reference film as the quality bar
- Measured pacing models and default voices for both Chinese and English keep final duration within ±15% of target
- One visual style by design (black-canvas MG); only the backdrop (stars vs dots) is a switch — anything else means editing style-guide.md and ui.tsx
- Narration is frozen once voiced: shot code hard-codes frame numbers, so a rewrite re-times everything
- Parallel builds are heavy: ~2–3 GB disk per film, ≥5 GB free required, and beyond ~12 agents you must dispatch in waves
- Windows untested (scripts developed and verified on macOS); no automated test suite
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Anything2Explainer this page | 56 · Use with care | ★ 3.1k | 1mo ago | NOASSERTION |
| PinMe Share | 54 · Use with care | ★ 3.8k | 1mo ago | MIT |
| Codeagent Multi-Backend Orchestration Skill | 51 · Use with care | ★ 2.8k | 6mo ago | AGPL-3.0 |
| Agent Skills Manager — manage-skills Skill | 50 · Use with care | ★ 43 | 7mo ago | MIT |
| 9Router Text-to-Speech | 46 · Use with care | ★ 32k | 9d ago | MIT |
Unlike generative video models (Sora, Veo, Runway) it produces deterministic code rather than synthesized pixels, and any frame can be fixed by editing one shot file. Unlike avatar tools (HeyGen, Synthesia) there is no presenter — motion-graphics diagrams show the mechanism. Unlike hand-rolled Remotion or Motion Canvas, it ships the full method (research → narration → storyboard → parallel build → QC) plus a reference film. Unlike Manim's Python math animations, it is an agent-driven React/TypeScript pipeline with TTS-aligned subtitles, chapters and quantitative QC.
How did FollowSkills review this skill?
Evidence shows four mandatory user confirmation checkpoints (length/language, narration sign-off, voiceover, 30s pilot), clear data-flow disclosure (edge-tts is a Microsoft cloud call, kokoro runs locally), royalty-free B-roll with MANIFEST logging, all frames drawn locally in code, explicit isolation of instructive text inside scraped research docs, no covert actions or external writes. Deducted: parallel builds consume 2–3GB disk with no stated rollback/cleanup path; edge-tts depends on a Microsoft endpoint; license metadata is NOASSERTION so attribution is incomplete. Not full marks.
SKILL.md is highly self-consistent: stages, scripts (tts_build/motion_check/selfcheck/frame_metrics/preview) and checkpoints cross-reference coherently; README corroborates and discloses known pitfalls (pin edge-tts 7.2.8, ARM install issues, pane caps). Deducted: static review finds no test suite or CI execution evidence; script bodies are not in evidence; failure-feedback quality on abnormal input rests on prose only. Capped at 10 by static calibration; scored 9.
Clear trigger semantics (user asks for an explainer video), explicit non-fit ranges (video replication, talking-head, live-action), bilingual support with per-tier length/resource budgets, thorough Linux/macOS environment notes. Deducted: heavyweight environment (Node, ffmpeg, venv, multi-agent parallelism, ≥5GB disk); mainland-China reachability of the Microsoft edge-tts endpoint and kokoro model downloads is not disclosed.
Excellent documentation layering (SKILL.md process + reference specs + lessons + FAQ + known limits), honest acknowledgements and originality statements, full PolyForm Noncommercial license and OFL font licenses. Deducted: repo metadata reads NOASSERTION vs the LICENSE text; no version numbers or changelog; maintenance responsibility and update path are only implicit in a personal repo; validated only on Claude Code/Codex.
The value claim is strong (topic in, finished MP4 with voiceover, subtitles and progress bar out) and the reference film's full paper trail plus contrast frames serve as an auditable yardstick with four cost-controlled checkpoints. Under static review nothing was executed: render quality, QC pass rates and efficiency claims (8 agents / 40 minutes) cannot be independently confirmed, exceeding the static cap of 7 is barred; the 1–3 hour / 2GB cost-benefit is unverified. Scored 6.
Evidence types are relatively rich: a complete reference-film paper trail (research/storyboard/QC reports/frames), 6 contrast frame pairs, two language cuts, and a lessons log distinguishing fact from design intent. Deducted: all material is author-supplied, with no third-party execution or CI+test coverage; the key reproducibility claim (deterministic frame-level rendering) cannot be statically verified. Scored 4.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →