split-screen
/split-screen
Split screen for a short - visuals on top, a host image below, captions on the seam - over one or more consecutive phases; the top visual swaps on spoken words. Use when the user asks for split screen, "host below", or a talking-host layout for a faceless short.
- Author
- aibeforeyouclick ↗
- Repository
- shorts-layouts ↗
- Licence
- MIT
- Last repository change
- 02.11.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 02.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Split screen for a short - visuals on top, a host image below, captions on the seam - over one or more consecutive phases; the top visual swaps on spoken words. Use when the user asks for split screen, "host below", or a talking-host layout for a faceless short.
Triggers from the description
Original author description
Split screen for a short - visuals on top, a host image below, captions on the seam - over one or more consecutive phases; the top visual swaps on spoken words. Use when the user asks for split screen, "host below", or a talking-host layout for a faceless short.
How it works
SKILL.md has no level-two headings · 13 lines, 347 words
SKILL.md preview
---
name: split-screen
description: Split screen for a short - visuals on top, a host image below, captions on the seam - over one or more consecutive phases; the top visual swaps on spoken words. Use when the user asks for split screen, "host below", or a talking-host layout for a faceless short.
---
# /split-screen
Spec: `docs/SPEC.md`. Example: `examples/layout-check-2` (phase p1).
1. Host: one image of your character in `examples/<slug>/assets/` (1080x960 works best; it slowly zooms so the bottom never sits still), or a built-in character: `"mascot": "lens"` (a lens drawn inline and driven by the word timings: idle = the eye floats, glances and blinks; speaking = 5 bars jump on every word with a pulse ring). Video hosts are not supported.
Dialogue: `"mascot": "lens+clicky"` puts Lens (left) and Clicky, a cursor character (right), in the bottom half. The script needs `"cast": {"lens": {"voice": "am_michael", "speed": 1.0}, "clicky": {"voice": "af_heart", "speed": 1.0}}` and lines written as `{"who": "clicky", "text": "...", "gap": 0.2}` (gap = pause after the line in seconds; negative = interruption). tools/tts.py trims each line's own silence and writes envelope.json, so the mouth and bars follow the real loudness frame by frame; the listener looks at the speaker; captions take the speaker's colour. Tune the comic timing with the per-line gaps, then run tools/sync_index.py before build_data. Voice engine: add `"engine": "chatterbox"` for Chatterbox-Turbo (more natural; cues `[laugh]` `[sigh]` `[chuckle]` inside a line's text; voices from voices/lens.wav and voices/clicky.wav, short clips of voices you have the rights to). Lines are cached in audio/lines/, so changing gaps re-times instantly; a bad take gets a new `"seed"` on that line. Word times come from whisper. Kokoro `[word](/phonemes/)` overrides are ignored by Chatterbox (it reads the plain word), so listen to heteronyms. Setup: README, "Dialogue".
2. Top visuals: 1 to 6 images (1080x960; the top 75 px and bottom 75 px get cropped). The first shows from the start of the split; each next one swaps in on its word.
3. Add to `script.json`:
`"split": {"phases": ["p1"], "host": "assets/host.svg", "top": [{"src": "assets/a.svg"}, {"src": "assets/b.svg", "anchor": "checklist", "n": 1, "phase": "p1"}]}`
4. Pipeline: check_script, tts, build_data, then `python -I tools/split_screen.py examples/<slug>`.
5. `npx --no-install hyperframes check`, snapshot at 0.05 s and after each swap; check the AI-GENERATED VOICE badge is clear of the top image and captions sit on the seam.
6. Report output paths and what was not checked. Never publish.
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · aibeforeyouclick/shorts-layouts | 96 | MIT |
| artemnovitckii/claude-editing-skills | 100 | MIT |
Database snapshot: 13.11.2026 · package version f159356c9a9b (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100short-video-formats
When the user wants to choose or plan the format of a promo short video for TikTok, Instagram Reels, YouTube Shorts or Xiaohongshu, especially for a software product, dev tool or AI agent. Covers the four formats that carry this category (talking head over a demo, phone filming… - 100spoken-captions
Put spoken words on screen from the transcript - word-timed captions on the user's own talking footage, and on a cut reel, captions per clip and a few highlight phrases timed to the words they call out, placed with scripts/reel_text.py. Use when a video is watched muted and its… - 100spoken-captions
Put spoken words on screen from the transcript - word-timed captions on a generated voice-over or on the user's own talking footage, and on a cut reel, captions per clip and a few highlight phrases timed to the words they call out, placed with scripts/reel_text.py. Use when a… - 100reel-split
Use when the user runs /reel-split, or asks for the split-screen explainer Reel — an animated widget panel on the top ~45 % of the frame, the speaker below it the whole time, no subtitles, no SFX — typical for "I found this tool / here are 5 tips" videos, on a talking-head take in raw/. - 100shortstock
Turn a script into a narrated stock-footage video (faceless video) with the shortstock CLI. Use when the user asks to make a voiceover video, a faceless short, a YouTube/TikTok explainer from a script, or to turn text into a video with stock footage and captions. - 100render-split-screen-creator
Assemble a split-screen creator ad from a config: an AI-creator lip-sync take fills the bottom half while real 16:9 product or demo clips play uncropped in the top half over a blurred fill of the same clip (never black bars), with a brand-colour divider between them. Scenes…