Independent by design · no pay-to-rank placements  · How we evaluate tools →
Anclum Submit
S

sabotage-dev/sabotage-motion

Independent directory — not the official site. Details below come from the project's public repository.

Motion graphics for short-form video, written in code by your coding agent.

Before you install: no copy-paste command · license MIT · link reachable

First step: nothing runnable here yet - start from the repository: https://github.com/sabotage-dev/sabotage-motion

updated 2026-11-18
View on GitHub ↗Link checked 2026-11-17 · HTTP 200 · opens the vendor's own site
Star count
0 ★ · sabotage-dev/sabotage-motion · read from the GitHub search API · (checked 2026-11-19 20:00:48)
Source
read directly from its GitHub page · type: open-source project
Harvest history
first seen 2026-11-15 · last seen 2026-11-20 · seen 6× · listed by 2 sources
Score
score 203 · 2 sources · repo
FAQ (answers come from our records)
Who is it for?
We filed it under these tags: Open Source. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository sabotage-dev/sabotage-motion;first catalogued 2026-11-15;last seen 2026-11-20。There is no official certification or quality rating here — we record sources and dates only.
README excerpt (English, untranslated) · Read the full README on GitHub →

# sabotage-motion

Excerpted from the repository README (abridged)

Motion graphics for short-form video, written in code by your coding agent.
A Claude Code skill. The agent writes a scene as a pure function of time, frame(t), drawn with Canvas2D over an
optional WebGL2 pass (bloom, grain, shader backgrounds, real motion blur). Headless Chrome renders every frame
exactly and pipes it to ffmpeg. Sound effects are synthesized in code on the same timeline, so they land on the
frame. No React, no framework, no render service: Node, Chrome and ffmpeg.
Planned for v1 (the storyboard, voice-driven timing and captions are built; mix and master and footage are built and in review):
Storyboard. One board.js per piece: each beat's voice line, on-screen copy, visual note and sound cue. A
board sheet shows a panel per beat, first as slates and then as real frames, and a slate scene gives you an
animatic with the real voice before anything is animated.
Voice-driven timing. Word or character timings from a TTS take or a transcript become cue points, so cuts
land on the words. A hook-length check is optional.
Captions. Word-by-word captions from the same timings.
Mix and master. Voice, an optional music bed ducked under it, and the synthesized effects, mastered to
-14 LUFS for social.
Footage. Screen recordings and clips as frame sequences, so they stay frame-exact.
check. Measured checks before you post: size, fps and frame count, loudness and true peak, stretches where
nothing moves, an empty first frame, and text outside the platform's safe zones.
Transparent output. ProRes 4444 or WebM with alpha, for overlays on other footage.
## Status
Early build, not ready for use yet. The v1 build plan is docs/PLAN.md. Phase P0 is in place: the
engine template with bundled open-licence fonts, --scene and --target, a render.json record after every
video render, doctor, and CI on macOS and Linux. Phase P1 is merged (PR #2): the storyboard (board.js, the board
sheet, the slate scene), voice-driven timing (tools/cues.mjs, with the hook gate), the optional ElevenLabs voice
tool, and captions. Phase P2 is in review: mix and master (tools/mix.mjs) and footage (tools/frames.mjs,
loadSeq and seqAt). check comes in P3, and the skill text (SKILL.md is a stub for now) in P4.
## Storyboard and voice (P1)
Each piece has a board.js: its beats in order, each with a voice line (vo) or a length (dur), the
on-screen copy, a visual note and a sound note. The scene builds its timeline from it (beatWindow(id),
beatAt(t), CFG.DUR = boardDur() in lib.js), so the board is the one place timing lives. Run these in the
piece's folder:
sh
node render.mjs board --slates # board.png and board.md: a placeholder card per beat, estimated timing
node tools/voice-elevenlabs.mjs --model eleven_v4 # optional: voice the board in one take (take-raw.wav, take.json)
node tools/cues.mjs # cues.js, cues.json and vo.wav from take.json and take-raw.wav
node tools/cues.mjs --from whisper read.json --audio read.wav # or: a human read, transcribed with openai-whisper
node render.mjs video --scene slate.js --out out/animatic.mp4 # the animatic's picture, timed by the take
node render.mjs board # the board sheet again, with the real scene and the take's timing
Listed badge (optional)
Copy this line into the README if you want the listing mark. Optional, and it does not affect being listed.
[![Listed on Anclum](https://anclum.com/badge/listed.svg)](https://anclum.com/en/s/sabotage-dev-sabotage-motion/)
Source
github · https://github.com/sabotage-dev/sabotage-motion
Browse by field
Open source
Listed by
2 independent sources
added
today

More entries

All 1,875 →

Filed under

Every tag opens the full list for that filter. Counts are the live catalogue figures.

Browse all fields →

← Back to index