Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.
- agent-skill
- video
- promo
- marketing
anything2explainer
Topic in, narrated explainer video out — a Claude Code / Codex skill that produces full motion-graphics explainers with Remotion, every frame drawn in code.
From the author of video-talkcraft: pick a topic, length, and language, and the agent researches it with sources, writes the narration, generates voiceover and a frame-accurate timeline, storyboards every shot, then dispatches parallel build agents that write one Remotion (React + TypeScript) component per shot. QC agents review rendered frames against written criteria. Output is a 1280×720 H.264 MP4 with word-boundary-aligned subtitles, chapter cards and a progress bar — no stock footage, no generative video model.
What it does
- Full film pipeline in one skill: research with sources → narration → voiceover + frame-accurate timeline → storyboard → parallel shot-building agents → QC against written criteria
- Every frame is a Remotion component (React + TS) on a black canvas — star-field or dot-field backdrop, white line art, ultra-bold headlines; no stock footage or gen-video models
- Chinese voiceover via edge-tts (zh-CN-YunxiNeural), English via local kokoro-82m, or bring your own TTS / finished audio with forced alignment
- Three length tiers (2–3 / 3–5 / 5–8 min) with a published table of narration budgets, shot counts, build-agent counts and wall clock (≈1–3 h); you are consulted at exactly four checkpoints
When to use it
When you need a reusable, brand-consistent explainer for a technical topic — vector databases, RAG, your product's architecture — and want the whole film reviewable as code instead of a timeline file.
Getting started
git clone https://github.com/Vincentwei1021/anything2explainer.git ln -s "$PWD/anything2explainer" ~/.claude/skills/anything2explainer # Claude Code ln -s "$PWD/anything2explainer" ~/.codex/skills/anything2explainer # Codex # deps: Node ≥18, ffmpeg, and a venv with edge-tts==7.2.8 + numpy pillow scipy # English narration additionally needs kokoro + soundfile + espeak-ng
Things to know
PolyForm Noncommercial license — fine for your own videos, not for commercial work. Heavy deps (Remotion, ffmpeg, pinned edge-tts); macOS is the verified platform (Linux should work, Windows untested), and a full film takes 1–3 hours of wall clock plus ≈2–3 GB of disk.
Repository
Vincentwei1021/anything2explainerCharted
Related on the map
geo-sleuth
SkillAn agent skill that geolocates a photo with nothing readable in it — OSM geometry, elevation skylines and satellite tiles turn “where was this taken?” into a measured answer with an error radius.
- agent-skill
- geolocation
- osint
- openstreetmap
Birdview
SkillA skill that makes AI coding agents map the architecture and declare the blast radius before they touch a line of code — rendered as one standalone HTML page.
- skill
- architecture
- code-visualization
- change-scope