- Home
- Skills
- Image & Video
- Cinematic Knowledge Shorts
Works with the AI Agents you already use
Cinematic Knowledge Shorts
Generate code-driven, cinematic educational videos with bilingual subtitles and synthesized BGM from a single fact.
Free
Cinematic Knowledge Shorts
Example session with this skill installed
Create a cinematic knowledge short about the '37% Rule' in optimal stopping theory. The author name is @MathLogic. Use the simulation grid visual motif.
- Read your context and instructions
- Compiled the cinematic knowledge shorts
- Generated the image
I have analyzed the 37% Rule (Secretary Problem) and drafted the production assets. I verified the mathematical constants ($1/e \approx 0.368$) and structured the narrative to transition from a "dating" metaphor to the formal probability proof.
Cinematic Production Plan: The 37% Rule
| Phase | Deliverable | Status |
|---|---|---|
| Script | script.md | 48 bilingual lines generated (02:15 duration) |
| Storyboard | storyboard.md | Defined "Linen-card" layout and "10,000-dot grid" simulation |
| Logic | index.html | JS Canvas code for real-time probability curve plotting |
| Audio | bgm.wav | 77 BPM Minimalist Piano track synthesized |
Visual Preview (Frame 00:42)
- Top Label: [ 概率与决策 / Probability & Decision ]
- Center: A grid of 100 gold and white nodes, simulating the rejection of the first 37 candidates.
- Annotation: "n/e critical threshold"
- Subtitle: 这里的关键是:先观察前37%,但不要做决定。 / The key here: observe the first 37%, but do not leap.
Code Snippet (Animation Core)
// index.html - Core Simulation Logic
window.render = function(t) {
ctx.fillStyle = '#050510';
ctx.fillRect(0, 0, 1920, 1080);
const progress = Math.min(t / duration, 1);
const threshold = 0.37;
// Draw simulation grid with fixed seed
drawAgentGrid(ctx, t, threshold);
// Update dual-language subtitles
const currentLine = script.find(l => t >= l.start && t < l.end);
if (currentLine) {
drawSubtitle(ctx, currentLine.cn, currentLine.en);
}
};
Next steps
- Review the
storyboard.mdto ensure the "Secretary Problem" metaphor meets your requirements. - Run
python3 scripts/render_frames.py index.htmlto generate the raw frames. - Finalize the build using
bash scripts/build_video.shto merge the BGM and visual sequence.
index.html
PNG · 1536×1024
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
What it does
Creating a cinematic explainer usually means editing software, stock footage or manual keyframing. This skill does it in code: you give it one fact, theorem or paper, and it produces a 2-minute text-driven knowledge short with a documentary aesthetic — every frame drawn by Canvas, the score synthesized, the subtitles burned in.
What you get
- 1920x1080, 30 fps, 2:00–2:45 MP4 (H.264)
- Bilingual subtitles — Chinese main line, English italic underneath
- Synthesized instrumental score, sectioned by a cue sheet
- A seven-step workflow: topic check → script → shot list → your confirmation → code → render → QA
Requirements
ffmpegwith H.264- Python 3 with
playwright,pillow,numpy, plusplaywright install chromium - Your own LLM API key — the skill calls your model, not ours
- Optional:
tesseract-ocr-chi-simfor subtitle QA - Windows: use
python, and runbuild_video.shfrom Git Bash
What it is not
No voiceover, no TTS, no stock footage, no AI image generation, no 4K. Every visual is drawn from scratch by a render(t) function, so nothing is licensed from a third party.
What it ships
Code, style references and prompts only. No third-party video, audio, image or transcript files are included.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.
No vertical 9:16 support. No built-in TTS voiceover. Requires local FFmpeg and Python environment for final rendering. Fact-checking relies on user-provided or web-searched data accuracy.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 17 days ago
- Free with an Agensi account
- Accounts
OpenAI
Anthropic
- Software
- create-video
Node.jsremotion
Python
Git
Permissions
Allowed Hosts
File Scopes
The skill renders frames locally through Playwright, writes them to disk, merges them with ffmpeg, and calls the buyer's own LLM endpoint to write the script and shot list. Nothing is uploaded to Agensi or to the creator; the only outbound traffic is the buyer's own model calls, authenticated with a key they set as an environment variable.