Animation Principles Master

This one treats animation principles as cognitive science rather than just art rules. The approach is refreshingly specific: squash and stretch aren't stylistic choices, they're matching what our brains expect from persistence of vision and observed physics. It frames staging as information architecture for human attention limits, and breaks down why anticipation makes motion readable to our predictive processing systems. If you're building animation tools or teaching Claude to critique motion work, the perceptual psychology angle gives you better reasoning foundations than the usual "make it look good" heuristics. The pose-to-pose versus straight-ahead framing as different philosophies is a nice touch for articulating intent.

GitHub stars: 70
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add dylantarre/animation-principles --skill 'Animation Principles - Master' --agent claude-code

Installs into .claude/skills of the current project.

Files

The Philosophy Behind the Principles

Masters don't just apply principles - they understand their origins in human perception and storytelling. This is animation as visual psychology.

Why the Principles Work

Squash and Stretch: Persistence of Vision

The eye blurs motion. Stretch simulates motion blur our brains expect. Squash simulates impact compression we've observed since infancy. We're matching internal neural predictions.

Anticipation: Predictive Processing

Human brains constantly predict next moments. Anticipation feeds this system information, making action legible. Without it, perception lags behind action.

Staging: Attentional Bandwidth

Humans have limited focal attention. Staging respects cognitive constraints. It's not artistic preference - it's information architecture for biological processors.

Method Philosophy

Pose-to-pose: Animation as acting - clear intentions, deliberate choices. Straight ahead: Animation as discovery - emergent behavior, organic evolution. Both philosophies are valid worldviews.

SKILL.md on GitHub