Vox AI Motion Graphics Generator skill
A skill that builds a narrated paper-collage explainer video from a topic, a talking-head clip or a single photo
- 281 ★ on the repository
- Last commit
Broad access
Why this level
- Sends your photos, presenter recordings and texts to the MuAPI cloud service, and voice cloning works from your voice sample
- Spends money on paid image, video and music model calls through your key
- Runs ffmpeg and Python scripts that write files into the `out/` folder
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/vox-ai-motion-graphics-generatorDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Sends your photos, presenter recordings and texts to the MuAPI cloud service, and voice cloning works from your voice sample
- Spends money on paid image, video and music model calls through your key
- Runs ffmpeg and Python scripts that write files into the `out/` folder
This is third-party code. Review the repository files before installing.
What it does
The `muapi-director` skill walks the agent through the whole production chain of a collage-style video: script, keyframes, animation, voice-over, music and captions. The result is a `final.mp4` file. You approve twice: the story beat map and the visual theme. There are three input modes. In B-roll the agent writes a hook-led story from a topic, draws each beat as a torn-paper poster and animates it. In A-roll a presenter recording is split into segments by speech recognition and redrawn as a collage while the face, gestures and lip-sync are preserved. In C-roll a selfie or product photo is cut out as a photographic sticker and anchored into animated posters. Before you pick a theme, the `style_bakeoff.py` script renders one beat in three or four styles. The cloud service MuAPI generates images, video, voice and music, while local ffmpeg handles assembly, ducking the music under the voice and burned-in captions. Every stage is a separate script in `scripts/`.
Who it is for. For explainer video makers and marketers who want a recognizable collage look without manual editing.
Good fit when
- You need a 15-second video on the history of coffee or another topic from a single request to the agent
- You want to redraw a speaker recording as a paper collage and keep their voice and expressions
- A product photo has to be placed into animated posters as a photographic sticker for an ad
- Before spending money on animation you want to compare one beat in several styles
Not a fit when
- You need to work without paid external services: all images, video, voice and music come from MuAPI through a key
- You need documentary footage from real filming: the skill produces only collage graphics
- Photos, voice and recordings of people must not go to a third party, since A-roll and C-roll upload material to the cloud
- You need a license: none is stated in the repository
Example request
Make a 15-second Vox-style collage video about how coffee reached Europe, in English, with music and captionsLimitations
You need keys for two services: `MUAPI_API_KEY` for image, video, voice and music models and `OPENAI_API_KEY` for story planning and speech recognition. Generation is paid and the repository gives no exact prices. `ffmpeg` with `ffprobe` and Python 3 with the `Pillow` package are also required.
The repository shows 5-second clips and the sample request in the README asks for 15 seconds. Animation comes from Veo 3.1 or Runway, so the look of the motion depends on the service. The repository states no license, and the Vox name refers to the style and implies no tie to the publication.
How to disable. Delete the skill folder from the agent's skills directory and remove the `MUAPI_API_KEY` and `OPENAI_API_KEY` keys from the environment. Finished videos sit in the `out/` folder next to the scripts.
FAQ
How do I install the Vox-style video skill in Claude Code?
Place the repository in Claude Code's skills directory and the agent picks it up from the `SKILL.md` file. Ask for a "vox video" or a "collage video". Codex and other agents read the instructions from `AGENTS.md`.
What are the B-roll, A-roll and C-roll modes?
B-roll makes a film from one topic. A-roll redraws a presenter recording as a collage while keeping the face and lips. C-roll anchors a photo of a person or product, cut out as a sticker, into the posters. Each mode has its own scripts, for example `aroll_clips.py` and `croll_keyframes.py`.
Which models does the pipeline use?
Frames come from `nano-banana-2` and `flux-dev`, animation from `veo3.1-image-to-video` or `runway-image-to-video`. Narration goes through `minimax-speech-2.6-turbo`, voice cloning through `minimax-voice-clone`, music through `suno-create-music`, and speech is recognized by `openai-whisper`.
How many decisions do I make myself?
Two: approve the story beat map and pick the visual theme. The agent does the rest automatically, from keyframes to the final file.
Where do I find the result?
For a project in `out/<project>` the pipeline creates `keyframes`, `clips` and `audio` subfolders, and the finished video is `final.mp4`. The `out/demo` sample is laid out the same way.
Discussion
No questions or reviews yet.
Related
SkillsAgentic video studio for Claude Code, Cursor and Codex: script, voiceover, images, stock footage, music, subtitles and render
Skills and CLI from HeyGen: the agent writes HTML compositions and renders them into MP4 videos, captions, motion graphics and decks
GPT-Image2 Style Library
gpt-image-2-style-library
A skill picks a matching template and style from a library of GPT-Image2 cases and assembles it into a finished prompt
Blender MCP
MCP for Blender
An MCP server and Blender addon: the agent creates and edits objects, materials and scenes, pulls assets and runs Python in Blender
A video editing skill for Claude Code and Codex: the agent cuts pauses and filler words, grades color, burns subtitles and adds animated overlays
An HTML design skill: clickable prototypes, slides with PPTX export, MP4 and GIF animations, infographics and expert review
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Vox AI Motion Graphics Generator (this entry) | no | 281 | ||
| anything2explainer | no | 2,710 | builds an animated explainer in Remotion code rather than collage posters through MuAPI | |
| short-video-maker | no | not required | 1,427 | an MCP server that assembles short vertical videos from ready materials and narration |
README in short
The README opens with the problem of scattered tools for explainer videos and presents the fix as a pipeline of screenwriter, collage artist, animator and editor. It then shows demos, key features, a model table, a quick start and the project structure.
README badge
Original SKILL.md
--- name: muapi-director description: > Turn ONE topic, talking-head video, or photo into a finished Vox-style paper-collage explainer / ad video on the MuAPI platform (api.muapi.ai) + local ffmpeg — script, collage keyframes, motion, voice-over, music, captions, all automated. Three input modalities: a topic (B-roll), a talking-head video (A-roll mode), or a single photo of a person/product anchored into the collage (C-roll mode). --- # MuAPI Director Turn a topic, talking-head video, or photo into a finished **Vox-style paper-collage video**: a bold, punchy, narrated explainer/ad where each beat is a torn-paper collage poster that comes alive, with voice-over, music and captions. Powered by **MuAPI** (`MUAPI_API_KEY`) + local **ffmpeg**. ## Input Modalities 1. **B-roll (topic → film):** Pass a topic (e.g. *"A brief history of Silicon Valley"*). It drafts a script, builds keyframe posters, animates them, synthesizes narration/BGM, and assembles the video. 2. **A-roll (talking-head video → collage):** Pass a presenter talking-head recording. It runs ASR, segments the transcript, and re-styles the video into paper-collage graphics while preserving facial features, lip-sync, and gestures frame-for-frame. 3. **C-roll (one photo → collage):** Pass a selfie or product shot. The subject is cut out as an unedited photographic sticker, anchored into generated collage posters per beat, and animated. ## Prerequisites - `MUAPI_API_KEY` set in your environment (get one from https://muapi.ai). - `ffmpeg` + `ffprobe` installed on system path. - Python 3 with `Pillow` (`pip install pillow`). ## Workflow Steps ### B-roll Mode (Topic) 1. Write `out/<project>/beats.json` containing topic, language, and beat map. 2. `python3 scripts/style_bakeoff.py out/<project>` (Pick theme from candidates). 3. `python3 scripts/keyframes.py out/<project>` (Generates collage posters). 4. `python3 scripts/clips.py out/<project>` (Animates posters into motion clips). 5. `python3 scripts/audio.py out/<project>` (Generates TTS narration and instrumental BGM). 6. `python3 scripts/assemble.py out/<project>` (Muxes audio, ducking, captions, watermark -> `final.mp4`). ### A-roll Mode (Talking Head) 1. `python3 scripts/asr_beats.py out/<project> <source.mp4>` 2. `python3 scripts/aroll_clips.py out/<project>` 3. `python3 scripts/aroll_assemble.py out/<project>` ### C-roll Mode (Photo Anchor) 1. Add `"mode": "croll"`, `"anchor_photo": "photo.png"`, `"croll_subject": "portrait"|"product"` to `beats.json`. 2. `python3 scripts/croll_keyframes.py out/<project>` 3. `python3 scripts/clips.py out/<project>` 4. `python3 scripts/audio.py out/<project>` 5. `python3 scripts/assemble.py out/<project>`