- Home
- Skills
- Image & Video
- Frames From Nothing
Frames From Nothing
Studio-grade motion graphics from pure code: software 3D engine, no browser, no After Effects, no stock
Free
Works with the AI Agents you already use
Frames From Nothing
Example session with this skill installed
Build a 3D motion graphics logo sting for my brand.
- Read your context and instructions
- Compiled the frames from nothing
A rendered MP4 with 3D geometry, extruded typography, synthesized sound design, measured loudness, and a verified download link.
Connects securely to your tools. The creator never sees your data.
About this skill
Every frame computed, nothing downloaded. This skill renders motion graphics with a software 3D engine written in Python: real polygon meshes, perspective projection, depth sorting, shading, and camera moves. It needs only Python, Pillow, numpy, and FFmpeg. It works where Playwright, After Effects, and stock libraries are unavailable.
Browser-based video engines fail in locked-down sandboxes. Template files need paid software. Stock footage needs licenses. This skill removes every external dependency: geometry, typography, lighting, and sound are all generated in code, so the output is fully owned and license-clean.
-
PIL paste requires integer coordinates; project and round them.
-
alpha_composite needs RGBA on RGBA; paste with a mask is different from alpha_composite and both have their place.
-
Guard every segment function against zero-length time ranges.
-
drawtext needs explicit fontfile and expansion=none; escape colons.
-
amix with mixed mono and stereo inputs collapses to mono; pan each input to stereo explicitly.
-
Image analysis tools often require files under the workspace, not /tmp.
-
A 900-frame 1080p render takes minutes, not hours, but still preview first.
-
The verified download link, first.
-
Scene table with timestamps.
-
Measured duration, resolution, loudness.
-
Honest notes: what is stylized, what the user should check.
-
The renderer source stays with the project so any scene can be re-rendered or re-timed later.
Software rendering: photorealistic ray tracing and live-action footage are out of scope. Complex scenes render in minutes, not seconds.
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 any payment.
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 forever
No account required to browse
Trust & safety
Security scan
Security scan passed today
- Free with an Agensi account
Requires