Skip to content

HeyChengC /vibe-knowledge-film

Agent skill for Douyin-style silent knowledge shorts, drawn frame by frame in code

About vibe-knowledge-film

This skill makes the silent knowledge shorts that went viral on Douyin: 2 to 5 minutes, 16:9, no voice, one line of subtitle while the screen acts it out. Give it a script or just a topic and it picks a look, writes the subtitles, plans a shot table and builds a code project. You get a sample of about 30 seconds to approve before it makes the full film. It's 16:9 only and the docs are Chinese.

Best for: Explainer creators who want Douyin-style silent knowledge shorts and are happy to review a code-rendered draft.

What it does

  • Two looks: Nocturne for hard, abstract topics and Fable for acting out psychology or behaviour
  • Breaks down four reference films sentence by sentence, with six ways to draw an idea
  • Writes a storyboard table, then draws every frame with Canvas and JavaScript
  • Renders through Playwright and your local Chrome, then mixes sound effects in with ffmpeg
  • Uses open-source fonts (SIL Open Font License), fetched on first run

Quickstart

Rewritten from the README. Check the repo for the latest steps.

  1. 1
    Put the folder in your agent's skills directory and keep the name, e.g. ~/.claude/skills/vibe-knowledge-film.
  2. 2
    Install the renderer: the command below, then python -m playwright install and python engine/get_fonts.py.
    pip install -r requirements.txt
  3. 3
    Make sure Google Chrome and ffmpeg are installed. The render script opens your local Chrome.
  4. 4
    Start a project outside the skill folder: the command below (C = Nocturne, B = Fable).
    python scripts/new_film.py <project-folder> --world C
  5. 5
    Render with the command below, or just hand your agent a topic and let it follow SKILL.md.
    python engine/render.py <project-folder> --serve

Models and languages

Interface and docs

  • Chinese

Alternatives

Other agent skills projects people compare with vibe-knowledge-film.

gnipbao

story-to-handdrawn-video

Agent skill that makes a hand-drawn 3:4 video from a story or images

Agent skills
  • OpenAI
2.2k GitHub stars+96/30d21 days agoMIT

renmengwen

MuseDock

Local workbench for AI explainer videos and whiteboard animations

Video framework
  • Doubao seed-audio-1.0
  • MiniMax TTS
  • MiMo TTS
  • +1
162 GitHub stars+3/30dtodayApache-2.0

calesthio

OpenMontage

Video studio that your AI coding assistant runs

Video framework
  • Google Veo
  • Kling
  • Seedance 2.0
  • +7
68k GitHub stars+5.9k/30dyesterdayAGPL-3.0

neilsonnn

image-blaster

Turn one image into a 3D set you can reuse across every shot

Agent skills
  • World Labs Marble
  • Nano Banana
  • GPT Image 2
  • +2
11k GitHub stars+2.8k/30d5 months agoMIT

Featured on OpenMicroDrama

Do you maintain vibe-knowledge-film? Add this badge to your README so English speakers can find our write-up.

Featured on OpenMicroDrama
markdown
[![Featured on OpenMicroDrama](https://openmicrodrama.com/badges/featured.svg)](https://openmicrodrama.com/projects/heychengc-vibe-knowledge-film)

The badge links to this page. Want your description changed? Email support@openmicrodrama.com.

Reviewed Oct 14, 2026. Approved from the discovery queue 2026-10-14. Engine and scripts are MIT; the filled-in briefs and two example films are study copies of other people's shorts, so don't publish them as your own. A same-named copy under Dave-cc was deleted. We write these descriptions ourselves. The repo's own docs are the final word.