Vibe Knowledge Film (narration-free explainer)
Builds 2–5 minute, 16:9, voiceover-free knowledge films where one line of subtitle is literally acted out on screen, rendered frame-by-frame with Canvas + Playwright + ffmpeg.
What it does
This skill walks you through producing a narration-free knowledge short film in the Douyin "Vibe知识大赏" style: no human voice, the viewer reads a single bottom subtitle while the screen performs that exact sentence.
- Define the thesis and build a fact table (years, names, numbers with sources)
- Find a layperson-recognizable object for every concept — dots, lines, circles and bars are banned
- Write the subtitle script together with a shot/storyboard table; no storyboard, no project scaffold
- Pick one of two looks (B fable / C nocturne) and lock color meaning for the whole film
- Hand-write
film.jsand render deterministically via Canvas + seek(t) + Playwright + ffmpeg - Inspect stills with
--sheet,--stills,--motionand iterate at least two revision rounds - Register event SFX or sync a BGM track, then pass a 9-point final QC table
Its core doctrine: visuals must explain, not decorate. Any imagery that would work for any script at all is treated as a failure — the skill even ships a documented failed version of its own film for contrast.
Who it's for
- Science/knowledge creators willing to draw visuals in code
- Makers of short-form video who don't want voiceover recording
- Educational content designers visualizing abstract concepts
- Users comfortable with Canvas/JS, Python and ffmpeg
Examples
- "Make me a Vibe知识大赏-style film about entropy" → thesis + fact table → three imagery directions → storyboard → 30-second head render → full film.
- Hand over an existing popular-science article: the skill flags sentences that cannot be drawn (no object, no verb), proposes rewrites, and converts the piece into subtitles plus a shot table.
- You have a botched earlier cut: it pulls frames, compares them against the subtitles, remakes the same segment and delivers an old-vs-new comparison image.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/HeyChengC/vibe-knowledge-film/HEAD/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on AI Skill Mart. Copy the SKILL.m folder from the GitHub repo HeyChengC/vibe-knowledge-film into my ~/.claude/skills/vibe-knowledge-film/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/HeyChengC/vibe-knowledge-film.git ~/.claude/skills/vibe-knowledge-film⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal.
- Create the skills folder:
mkdir -p ~/.claude/skills - Clone the repo:
git clone https://github.com/HeyChengC/vibe-knowledge-film.git ~/.claude/skills/vibe-knowledge-film - Install prerequisites: Python 3, Node,
ffmpeg, and Playwright (pip install playwright && playwright install chromium). - If fonts are missing, run
python ~/.claude/skills/vibe-knowledge-film/engine/get_fonts.py. - Restart Claude Code, then ask for something like "make a narration-free subtitle explainer video" to trigger the skill.
- Rendering is slow — validate the first 30 seconds (
--clip 0 30 head) before committing to the full film.