Visualize Skill: Interactive 3D Math & Science Visualizer

SkillWeb & browsing

Builds interactive, browser-based 3D Three.js visualizations of any mathematics or science concept as a single, self-contained HTML file. Takes a concept and a depth level (Simple, Intermediate, Advanced). Trigger on "ape visualize", "ape visualize math", "ape visualize science", "visualize math", "visualize science", "visualize concept", "ape simulate", or "/ape-visualize".

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Visualize Skill: Interactive 3D Math & Science Visualizer skill

What this skill tells your AI

The instructions your AI receives, as published by arpitbbhayani/ape-skills in ape-visualize/SKILL.md and read by ahel’s review.

Takes any concept from mathematics or science and creates an interactive, visually stunning, browser-based 3D visualization using Three.js, packaged as a single self-contained HTML file.

The visualization is grounded in real mathematical formulations and physical laws—not just decorative graphics, but a live, interactive simulation or geometric construction that users can rotate, zoom, tweak with sliders, and inspect in real time.


Inputs

The user provides:

  1. Concept: Any mathematics or science concept. For example:
    • Math: Eigenvalues & Eigenvectors, 3D Linear Transformations, Gradient Descent on Surfaces, Curl & Divergence of Vector Fields, Riemann Surfaces, Möbius Strip & Topology, Fourier Epicycles in 3D, Lorenz Attractor & Chaos, Spherical Harmonics, Minimal Surfaces (Helicoid/Catenoid).
    • Science: Lorentz Force, Gravitational Orbits & Precession, Double-Slit Wave Interference, Maxwell-Boltzmann Distribution, VSEPR Molecular Geometry, Quantum Harmonic Oscillator, DNA Replication Fork.
  2. Depth: One of three simple depth levels (default to Intermediate if not specified):
    • Simple (Level 1 — Intuition & Big Picture)
    • Intermediate (Level 2 — Mechanics & Equations)
    • Advanced (Level 3 — Deep Dive & Rigorous Internals)

Regardless of the starting depth, every generated HTML document embeds an in-browser Depth Selector (Simple | Intermediate | Advanced) so the user can fluidly toggle between levels live.


The Three Depth Levels (Simple Words)

The depth determines both the visual representation and the level of mathematical/physical rigor exposed:

DepthTarget AudienceCore FocusVisual Elements (Math & Science)Interactive ControlsTelemetry & HUD
SimpleBeginners, general audience, visual learnersBig-picture intuition & visual metaphors. What does it look like? What is the main effect?Math: Deforming shapes, flowing streamlines, animated paths, color gradients.Science: Macro objects, smooth orbits, glowing wavefronts. No overwhelming formulas.Play / Pause, Speed slider, 1–2 intuitive dials ("Morph", "Stretch", "Strength", "More / Less", scenario presets)."What's happening?" card in plain English explaining the core takeaway and visual intuition.
IntermediateUndergraduates, engineers, technical learnersHow it works mechanically — governing rules, vectors, coordinates, and equations.Math: Basis vectors ($\mathbf{\hat{i}}, \mathbf{\hat{j}}, \mathbf{\hat{k}}$), tangent planes, 3D vector arrow grids, coordinate axes, parameter curves.Science: Force/velocity arrows, field lines, particle trails, component breakdowns.Sliders with real variables and units (matrix components $a_{ij}$, surface curvature $k$, mass $m$, charge $q$, frequency $\omega$).Live numerical telemetry (current coordinates, matrix determinant $\det(A)$, energy, flux) + live 2D canvas sparkline. Typeset formula.
AdvancedResearchers, math/physics majors, deep diversRigorous math & deep mechanics — differential equations, phase space, edge cases.Math: Invariant eigenvector axes, differential forms, phase portraits ($\dot{x}$ vs $x$ or complex plane), branch cuts, geodesic curvature.Science: Runge-Kutta trajectories, Hamiltonian drift, quantum probability clouds, perturbation indicators.Precision mathematical dials (eigenvalue parameter $\lambda$, perturbation $\epsilon$, integration step $\Delta t$, non-linear coupling, boundary conditions).Real-time phase-space canvas, conservation/drift monitor ($\Delta E / E_0$, $\det(A)$ conservation), matrix/tensor readout, raycaster probe on click.

Output Requirements

Always output a single, self-contained HTML file (<concept>-visualization.html) that works immediately upon opening in any web browser without any local server or build tools needed.

Architecture of the HTML File

  1. Zero build step & Zero CORS: Loads Three.js and OrbitControls via reliable CDN <script> tags using standard UMD format:
    <script src="https://cdnjs.cloudflare.com/ajax/libs/three.js/r128/three.min.js"></script>
    <script src="https://cdn.jsdelivr.net/npm/three@0.128.0/examples/js/controls/OrbitControls.js"></script>
    
  2. Modern Scientific & Mathematical Dark UI: Clean OLED-dark glassmorphic design (#090a0f background, frosted glass panels with backdrop-filter: blur(14px), subtle neon accents: Cyan #00f0ff, Mint #00ff9d, Amber #ffb700, Coral #ff3366, Violet #b388ff).
  3. HUD Components:
    • Header Bar: Concept title, domain badge (MATHEMATICS, PHYSICS, etc.), and the 3-level Depth Switcher (Simple | Intermediate | Advanced).
    • Main 3D Viewport: Responsive Three.js canvas with smooth orbit, pan, and zoom (THREE.OrbitControls with enableDamping = true).
    • Controls Panel (floating glass card):
      • Timeline / Animation: Play / Pause, Reset, Speed slider (0.1x to 3x).
      • Concept parameters tailored to the active depth.
      • Visual toggles: Coordinate Grid, Basis / Field Vectors, Trails / Curves, Labels, Tangent / Normal indicators.
      • Presets dropdown (e.g. "Identity", "Shear", "Pure Rotation", "Degenerate / Singular" for math; "Default", "Resonance", "Chaos" for science).
    • Live Telemetry & Graph Panel:
      • Live mathematical/physical readouts (determinant, eigenvalues, coordinates, velocity, energy).
      • Real-time 2D Canvas plot (sparkline, phase portrait, or function curve).
    • Explanation & Equation Card:
      • Intuition: Plain English summary of what is visible.
      • Governing Equation: Beautifully typeset formula (e.g. $A \mathbf{v} = \lambda \mathbf{v}$, or $\nabla \times \mathbf{F}$, or $\mathbf{F} = q(\mathbf{E} + \mathbf{v} \times \mathbf{B})$).
      • Interactive Probe: Raycaster to click/hover 3D elements and inspect their mathematical state.

Domain Coverage: Mathematics & Science

1. Pure & Applied Mathematics

  • Linear Algebra:

    • Matrix Transformations: Visualizing $T(\mathbf{x}) = A \mathbf{x}$ acting on a unit cube or unit sphere.
    • Eigenvalues & Eigenvectors: Visualizing lines that do not change direction under transformation, only stretch by $\lambda$.
    • Determinant: Geometric interpretation as the signed volume scaling factor of parallelpipeds.
    • Singular Value Decomposition (SVD): Factoring $A = U \Sigma V^T$ as Rotation $\to$ Scaling $\to$ Rotation.
    • Quadratic Forms & Conic Sections: Visualizing $\mathbf{x}^T A \mathbf{x} = c$ ellipsoids, hyperboloids, and paraboloids.
  • Multivariable Calculus & Differential Geometry:

    • 3D Scalar Surfaces & Tangent Planes: $z = f(x, y)$ with tangent plane and normal vector $\mathbf{n} = \langle -f_x, -f_y, 1 \rangle$.
    • Gradient Descent: A particle rolling down the steepest descent path $-\nabla f(x, y)$ on a terrain.
    • Vector Fields, Divergence & Curl: 3D arrow fields with micro-paddle wheels showing curl ($\nabla \times \mathbf{F}$) and flux expanding from sources showing divergence ($\nabla \cdot \mathbf{F}$).
    • Minimal Surfaces: Catenoid, Helicoid, and Enneper's surface (surfaces with zero mean curvature $H = 0$).
    • Curvature: Gaussian curvature $K = \kappa_1 \kappa_2$ and principal curvature directions on 3D manifolds.
  • Complex Analysis & Topology:

    • Riemann Surfaces: Multi-sheeted geometric representations of multi-valued complex functions ($w = \sqrt{z}$, $w = \ln z$).
    • Conformal Mappings: Grid lines in the complex plane mapped by $f(z) = z^2$ or $f(z) = e^z$, proving angle preservation.
    • Möbius Strip & Klein Bottle: Non-orientable surfaces, normal vector traversal returning flipped.
    • Torus Knots & Stereographic Projections: Projecting 4D spheres / Clifford Torus down to 3D space.
  • Dynamical Systems & Chaos:

    • Lorenz Attractor: Butterfly-shaped strange attractor showing sensitive dependence on initial conditions.
    • Phase-Space Portraits: Plotting $(\theta, \dot{\theta})$ for non-linear pendulums, limit cycles, and strange attractors.
    • 3D Fourier Epicycles: Epicyclic rotating phasor arms in 3D tracing arbitrary complex curves or knots.

2. Physical & Natural Sciences

  • Physics:

    • Electromagnetism: Lorentz force ($q(\mathbf{E} + \mathbf{v} \times \mathbf{B})$), cyclotron motion, magnetic dipoles, Biot-Savart induction.
    • Classical & Celestial Mechanics: Keplerian planetary orbits, Lagrange points ($L_1-L_5$), chaotic 3-body gravitational interactions.
    • Wave Mechanics & Optics: Wavefront superposition, double-slit interference, thin-film diffraction, Snell's law refraction.
    • Thermodynamics: Maxwell-Boltzmann velocity distribution of colliding particles, Brownian motion.
    • Quantum Mechanics: Wave packet dispersion, harmonic oscillator probability density, spin precession in $B$-fields.
  • Chemistry & Molecular Biology:

    • VSEPR Theory: 3D electron pair repulsion geometries (linear, tetrahedral, octahedral, trigonal bipyramidal).
    • Crystal Lattices: Unit cells (FCC, BCC, Simple Cubic, Diamond) with atom packing fractions.
    • DNA & Cellular Dynamics: Double-helix transcription fork, ion gradients in action potential propagation.

Step-by-Step Generation Workflow

When the user requests a visualization:

Step 1: Mathematical & Scientific Decomposition

  1. Parse the concept and target depth (Simple, Intermediate, or Advanced).
  2. Identify the core mathematical/physical invariants:
    • For Math: Invariant axes (eigenvectors), volume scaling ($\det A$), conservation of topological genus, orthogonality, gradient direction.
    • For Science: Conservation of energy/momentum, flux continuity, wave phase relationships.
  3. Design how the 3 depth levels map to this concept:
    • Simple: Pure visual intuition, intuitive sliders, macro visual takeaway.
    • Intermediate: Explicit equations, vector arrows, coordinate grid, real mathematical variables.
    • Advanced: Differential equations, phase portraits, invariant monitors, numerical diagnostics.

Step 2: Implement the Three.js Geometry & Simulation

  1. Mathematical Representation:
    • For surfaces: Use THREE.ParametricGeometry or dynamically updated THREE.PlaneGeometry with vertex height displacement.
    • For vector fields: Use THREE.InstancedMesh with arrow geometries or arrays of THREE.ArrowHelper for efficiency.
    • For curves & trajectories: Use dynamic THREE.BufferGeometry with pre-allocated array buffers.
  2. Animation / Numerical Update Loop:
    • Discrete updates per frame with clamped delta time (Math.min(clock.getDelta(), 0.05)).
    • Pre-allocate scratch vectors (const _v1 = new THREE.Vector3()) outside the render loop to prevent garbage collection stutter.

Step 3: Wire Up Reactive Controls & HUD

  1. Connect sliders to mathematical/physical state variables without recreating the 3D scene.
  2. Provide presets so users can jump to famous cases (e.g. for linear algebra: Shear, Reflection, Rotation, Projection / Singular).
  3. Connect the live 2D canvas to plot real-time telemetry (sparkline of coordinates or phase-space trajectory).
  4. Implement the raycaster to allow clicking on 3D elements to inspect local values.

Step 4: Write & Deliver the Single HTML File

Write the complete, self-contained HTML file directly to the workspace or artifact directory, ready to be double-clicked and viewed in any browser.

Signals

GitHub stars
43
Forks
3
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
ape-visualize
Source
github.com/arpitbbhayani/ape-skills