vox-explainer skill
Claude Code skill that turns one topic into a 60-second Vox-style explainer: script, keyframes, animation, voiceover and music.
- 155 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Texts and prompts go to an external generation API, and each run spends money on the user's key.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/vox-explainerDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Texts and prompts go to an external generation API, and each run spends money on the user's key.
This is third-party code. Review the repository files before installing.
What it does
The `vox-explainer` skill runs a topic through six stages: a beat script, voiceover, keyframes, animation, music and ffmpeg assembly. The narration is generated and measured first, so beat durations become the master clock for every clip. Generation goes through the unified Atlas Cloud API with a single `ATLASCLOUD_API_KEY`: xAI TTS for voice, Seedream for frames, Gemini Omni Flash for animation, MiniMax for music. State lives in project files, so one beat can be regenerated without touching the rest.
Who it is for. Creators of educational channels who need a short narrated explainer without hand editing.
Good fit when
- A one-minute social video about an event or topic with subtitles and music is needed.
- One frame is off and gets regenerated without rebuilding the rest.
- Before paid animation you want a cost quote and a cheap draft pass.
- A consistent visual style is held through an anchor frame and reference edits.
Not a fit when
- You need footage of real people or your own face and voice.
- Scripts and prompts cannot be sent to an external generation service.
- You want a long film; the skill targets a roughly one-minute explainer.
Example request
Make a 60-second Vox-style explainer about the Tang Dynasty golden age. Quote the cost first and offer a draft pass.Limitations
Every generation is paid: the README estimates about $3–8 per 60 seconds at July 2026 Atlas Cloud prices, mostly animation. You need `ffmpeg` and `ffprobe` with libass, Python 3.10+ and the `requests` package. The script and prompts go to Atlas Cloud and the models behind it. The skill imitates the manner of Vox; it does not copy branded material. Model versions and prices in the README are tied to July 2026 and may change.
How to disable. Delete the `~/.claude/skills/vox-explainer` folder and remove `ATLASCLOUD_API_KEY` from the environment. Finished video projects live separately and are deleted by hand.
FAQ
Which models are used?
xAI TTS makes the voice, Seedream 5.0 Pro the frames, Gemini Omni Flash the animation and MiniMax Music 2.6 the music, all called through Atlas Cloud.
What is in the repository?
A `SKILL.md` with the pipeline and three references in `references/`: Vox style, the Atlas Cloud API and ffmpeg assembly.
Why is the voice generated first?
Speech duration sets the timing of every beat, so clips are conformed to narration instead of squeezing speech into clips.
Can part of the video be rebuilt?
Yes. All progress sits in `script.json`, `vo_durations.json` and per-beat files, and any beat can be regenerated alone.
What must be installed first?
An `ATLASCLOUD_API_KEY`, `ffmpeg` and `ffprobe` with libass, and Python 3.10 or newer with the `requests` package.
Discussion
No questions or reviews yet.
Related
SkillsA skill that turns one topic into a finished Vox-style paper-collage video: script, keyframes, motion, voice-over, music and captions
A skill that builds a narrated paper-collage explainer video from a topic, a talking-head clip or a single photo
The video-explainer skill turns a script into a vertical 9:16 explainer: narration, captions, Remotion and a pre-render review
A Claude Code and Codex skill that turns a topic into a narrated motion-graphics explainer video on Remotion with subtitles and chapters
A skill that turns a topic into a narrated 4K video: research, script, local TTS, Remotion composition and an FFmpeg mix
A skill and CLI: the agent writes a script and scenes, while the voice, word timing and MP4 render run on your computer with no API keys
README in short
The skill turns one topic into a film through script, voiceover, keyframes, animation, music and assembly. Assembly runs locally with ffmpeg.