- Home
- Skills
- Game Development
- deterministic demo video
deterministic demo video
Automated, frame-accurate demo video rendering for headless simulations and game loops.
Free
Works with the AI Agents you already use
deterministic demo video
Example session with this skill installed
I need to render a 10-second demo of my canvas-based pulse animation at 60fps and verify that the audio onsets match the visual pulses. Generate the config and show the render command.
- Read your context and instructions
- Compiled the deterministic demo video
I have generated the configuration for your 10-second pulse animation. This configuration includes a timeline of events and the necessary verification probes to ensure audio-visual synchronization.
Demo Configuration
{
"adapter": "./examples/pulse/adapter.mjs",
"outDir": "./out/pulse-demo",
"settings": {
"fps": 60,
"duration": 10.0,
"width": 1280,
"height": 720
},
"timeline": [
{ "time": 0.0, "type": "start" },
{ "time": 2.5, "type": "pulse", "params": { "frequency": 440 } },
{ "time": 5.0, "type": "pulse", "params": { "frequency": 880 } },
{ "time": 7.5, "type": "pulse", "params": { "frequency": 440 } }
],
"expect": {
"onsets": [2.5, 5.0, 7.5],
"probes": [
{ "time": 2.5, "x": 640, "y": 360, "color": "#FF0000" }
]
}
}
To render the video, run the following command in your terminal:
node scripts/render.mjs --config pulse-config.json
After rendering is complete, you can verify the output accuracy using:
node scripts/verify.mjs --dir out/pulse-demo
Next steps
- Verify your
adapter.mjsimplements thereadFrame()andreadAudio()methods correctly. - Ensure
ffmpegis accessible in your current environment path. - Check the
manifest.jsonin the output directory for detailed render metadata.
Connects securely to your tools. The creator never sees your data.
About this skill
Screen recording is unreliable for automated pipelines because it drops frames, depends on window focus, and produces inconsistent results across runs. This skill replaces manual capture by driving your simulation or game loop directly, ensuring frame-accurate synchronization between audio and video via a two-pass deterministic rendering process.
What it does
- Deterministic rendering Ensures the same input timeline always produces a bit-identical or functionally identical output.
- Frame-accurate sync Uses content-time emission logic to prevent audio drift over long durations.
- Two-pass capture Captures audio and video in separate runs using a shared clock to maintain perfect alignment.
- Automated verification Measures the final encoded MP4 for frame count, duration, and audio onset accuracy.
- Headless support Works with WASM modules, Node.js scripts, Canvas animations, and game loops.
How it works
- Implement the adapter Create a script following the
adapter-api.mdcontract to bridge your simulation to the renderer. - Configure the run Define the event timeline, duration, FPS, and verification probes in a
config.jsonfile. - Execute the render Run the
render.mjsscript to generate the audio, silent video, and final muxed MP4. - Run verification Use
verify.mjsto validate the output against your expectations and catch regression errors.
Frameworks & tools
This skill requires Node.js 18+ and ffmpeg. It is designed to interface with simulations built in WASM, Canvas, or generic JavaScript game loops.
Why this beats prompting it yourself
Writing a two-pass renderer from scratch usually leads to tiny rounding errors that cause audio to drift out of sync after 30 seconds. This skill implements specific invariants for content-time frame emission and instance isolation that are difficult to get right through trial and error.
Use cases
- Generating product demo videos automatically during CI/CD release pipelines.
- Producing high-quality, perfectly synced visualizers for WebAudio or generative music.
- Creating reproducible bug reports for visual simulations or game mechanics.
- Batch rendering marketing assets from a headless canvas engine.
Known limitations
Does not support live screen capture or manual human-in-the-loop interaction. Requires a programmatic adapter for the target simulation.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Game Development.
Requires an adapter implementation for each project. Not for live screen capture. Needs ffmpeg installed on the system or via environment variables.
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 3 days ago
- Free with an Agensi account
- Software
Node.jsFFmpeg