zooming
Skill0 starsChanged 21 days agoMITView on GitHub
Video effect - an exponential camera push into one point on a screen (an icon, a mascot's eye), through into a tunnel of flying code, then a pull back out. Use when the user types /zooming or wants a "zoom into the screen / into the eye" transition in their footage.
In this repository8 other files it gives its agents.
- icon-on-palmSkill
- orbitSkill
- paperSkill
- share-commentSkill
- thanos-snapSkill
- visuals-coreSkill
- words-from-behindSkill
- zig-zag-graphSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 42
Put your own lines in --code (the reel's commands, your repo name) for a personal touch.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/zooming
The camera dives into a single point on the laptop screen, faster and faster, the pixels give way to a dark tunnel of glowing code lines (with an optional spinning icon in the middle), it holds there, then pulls all the way back out to the wide shot. Default timing fits ~3.6 s (1.15 s in, hold, 1.15 s out).
Needs: a locked-off clip, ideally 4K (the push renders from the source's full resolution, so 4K stays sharp much
deeper). Requires the visuals-core folder next to this one.
Steps
- Pick the target with
python ../visuals-core/grid.py --in clip.mp4 --t <start> --out grid.jpg, then--croparound the screen for precision. Dark targets (an eye, a pupil) make the best dive. - Preview the push, the blend and the tunnel, then render:
python zooming.py --in clip.mp4 --start 46.2 --dur 3.6 --out preview --preview 0.4,0.6,0.8,1.2,3.0 \ --target 894.6,1170.2 --center-icon logo.svg python zooming.py --in clip.mp4 --start 46.2 --dur 3.6 --out zoom.mp4 --target 894.6,1170.2 --center-icon logo.svg
Flags
| flag | default | meaning |
--target x,y | required | point to dive into (output pixels) |
--push | 1.15 | seconds to push in, and to pull out |
--max-zoom | 2400 | zoom factor at the deepest point |
--tunnel-from | 60 | zoom factor where the tunnel starts fading in (fully in at 10x) |
--code file.txt | built-in | one code line per text line, shown in the tunnel |
--center-icon | none | PNG/SVG spinning in the tunnel centre |
--glow | #D97757 | glow colour behind the centre icon |
Check before delivering
- At ~0.6 s the target is still recognisable and sharp (a 1080p source goes soft earlier; lower
--tunnel-from). - The pull-out lands exactly on the original wide shot.
- Put your own lines in
--code(the reel's commands, your repo name) for a personal touch.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

