Webgl Card Effects logo

Webgl Card Effects

Community
notque
webgl-card-effects

Standalone WebGL fragment shaders for card visual effects: holographic foil, shimmer, rarity glow.

Overview

Publishernotque
Repositoryvexjoy-agent
Skill namewebgl-card-effects
Stars
443
Forks
50
Bundled files
3
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by notque on GitHub. Read the source before you install it.

Installation

Install the Webgl Card Effects AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/notque/vexjoy-agent/tree/main/skills/frontend/webgl-card-effects
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/notque/vexjoy-agent.git /tmp/vexjoy-agent
mkdir -p .claude/skills
cp -r /tmp/vexjoy-agent/skills/frontend/webgl-card-effects .claude/skills/webgl-card-effects
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Webgl Card Effects in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Webgl Card Effects on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Webgl Card Effects is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

WebGL Card Effects Skill

Overview

This skill adds GPU-accelerated visual effects to React card components using standalone WebGL2 fragment shaders — no Three.js, no R3F, no external library required. It targets deckbuilder games and card UIs where rarity tiers should feel visually distinct, not just differentiated by a CSS box-shadow value.

Scope: Holographic foil overlays, metallic shimmer bands, rarity-driven energy pulses, and interactive tilt-shine effects mounted directly on React card components. The canonical target is FramedCard.tsx in a React 19 / Vite / Tailwind project with a rarity system (starter → common → uncommon → rare → legendary).

Not in scope: 3D transformations, texture loading, post-processing pipelines, or any Three.js scene management. For those, use threejs-builder.

Key constraint: Browsers cap WebGL contexts at roughly 8–16 total per page. This skill uses a single shared WebGL2 context with blit-to-2D-canvas output per card, avoiding the per-card context problem entirely. See references/shader-integration-react.md for the full singleton pattern.


Instructions

Phase 1: ASSESS

Goal: Understand which effects are needed and confirm the card component structure before writing any code.

Step 1: Read the card component

Read these files before writing anything:

  • src/components/cards/FramedCard.tsx — component structure, hover state, rarity prop flow
  • src/components/cards/cardStyles.ts — CARD_SIZE_CONFIG for pixel dimensions, rarity string values

Confirm:

  • How rarity reaches the component (prop name, TypeScript type, exact string values)
  • Whether isHovered state already exists in the component
  • Which sizes are rendered in contexts that justify shader cost (xl/lg only — xs/sm are too small)
  • Whether showShine / card-shine CSS exists and needs coordination

Step 2: Select effect tier per rarity

RarityWebGL?Effect
starter / commonNoCSS shimmer only — no WebGL overhead
uncommonYes (subtle)Metallic band shimmer, very low opacity
rareYes (medium)Moving shimmer + blue hue shift + edge pulse
legendaryYes (full)Rainbow holographic foil, mouse-reactive tilt

Step 3: Confirm React version

Check package.json for the React version. This skill assumes React 19 (ref as prop, no forwardRef). If the project uses React 18, canvas refs require useRef + standard ref passing.

Gate: Rarity values confirmed, CARD_SIZE_CONFIG read, effect tiers decided. Proceed only when this gate passes.


Phase 2: BUILD

Goal: Write working GLSL shaders and the WebGL initialization harness.

Load references/card-shader-patterns.md now.

Step 1: Create the shader strings module

Create src/components/cards/effects/cardShaders.ts. This file holds:

  • The shared vertex shader (passthrough UV coordinates)
  • Three fragment shader strings: SHIMMER_FRAG, RARE_FRAG, LEGENDARY_FRAG
  • A rarityToUniform(rarity: string): number mapping function

Every shader must expose this exact uniform interface:

glsl
uniform float u_time;        // seconds elapsed, JS wraps at 1000.0
uniform float u_rarity;      // 0.0=starter/common, 0.25=uncommon, 0.5=rare, 1.0=legendary
uniform float u_hover;       // 0.0 to 1.0, lerped by JavaScript each frame
uniform vec2  u_mouse;       // normalized card-space [0,1] mouse position
uniform vec2  u_resolution;  // canvas pixel dimensions (width, height)
uniform float u_upgraded;    // 0.0 or 1.0 — upgraded cards get slightly more intense effect

Step 2: Create the WebGL harness hook

Create src/components/cards/effects/useCardShader.ts.

Load references/shader-integration-react.md for the full hook source. The hook must:

  • Use the shared WebGL2 context singleton (not create a new context per card)
  • Accept { rarity, isHovered, isUpgraded, enabled } as input
  • Return a RefObject<HTMLCanvasElement> that the component attaches to the canvas element
  • Pause the animation loop when the card is off-screen (IntersectionObserver)
  • Run at 30fps maximum via delta-time throttle
  • Clean up the RAF loop and observer on unmount

Step 3: Shader construction for each tier

Load references/balatro-shader-breakdown.md for the legendary holographic shader GLSL source.

For rare: use the shimmer band + hue shift layer from the breakdown, omit the rainbow foil layer.

For uncommon: use only the metallic band pass (single moving highlight), opacity 0.3 maximum.

Gate: Run npx tsc --noEmit. Zero TypeScript errors. Open browser console and verify gl.getShaderInfoLog() returns empty string for all shaders.


Phase 3: INTEGRATE

Goal: Mount the canvas overlay on FramedCard.tsx and wire rarity/hover state into the shader uniforms.

Step 1: Derive render decision

Inside FramedCard, after the existing const shouldShine line:

tsx
const shouldRenderShader =
  ['uncommon', 'rare', 'legendary'].includes(rarity) &&
  size !== 'xs' &&
  size !== 'sm';

Step 2: Call the hook

tsx
const shaderCanvasRef = useCardShader({
  rarity,
  isHovered,
  isUpgraded,
  enabled: shouldRenderShader,
});

Step 3: Add the canvas overlay to JSX

Inside the motion.div return, immediately after the frame <img> element (after z-10):

tsx
{shouldRenderShader && (
  <canvas
    ref={shaderCanvasRef}
    className="absolute inset-0 w-full h-full pointer-events-none rounded-lg"
    style={{ zIndex: 15, mixBlendMode: 'screen' }}
  />
)}

mix-blend-mode: screen makes the shader's black background transparent while letting bright holographic colors add onto the card surface.

Step 4: Coordinate with existing CSS shine

The existing card-shine CSS class creates a gradient sweep on hover. It will double-shimmer with the WebGL effect. Suppress it for rarities that have the WebGL shader:

tsx
const shineClass =
  showShine && shouldShine && !shouldRenderShader
    ? `card-shine ${...}`
    : '';

Step 5: Verify z-layer stack

From bottom to top inside the card:

  • z-0 — artwork container
  • z-10 — frame PNG image
  • z-15 — shader canvas (new)
  • z-20 — text elements (energy orb, name, description, type strip)
  • Tooltip renders outside via AnimatePresence portal

Tailwind does not generate z-15 by default. Either add it to tailwind.config or use style={{ zIndex: 15 }} inline (already shown above).

Gate: Cards render correctly at all sizes. Shader canvas is visible on uncommon/rare/legendary. No z-fighting between shader layer and frame image. TypeScript clean.


Phase 4: POLISH

Goal: Performance verification, visual tuning, mobile fallback confirmation.

Step 1: Performance audit

Open Chrome DevTools → Performance tab. Record 5 seconds while hovering over a legendary card.

Targets:

  • GPU usage: < 5% for a single legendary card
  • Frame time: shader must not push game loop below 60fps
  • No memory growth across multiple mount/unmount cycles (check Heap in Memory tab)

Step 2: Mobile fallback verification

typescript
// In useCardShader.ts — call this once at module load
function supportsWebGL2(): boolean {
  try {
    const canvas = document.createElement('canvas');
    return !!canvas.getContext('webgl2');
  } catch {
    return false;
  }
}

On devices where supportsWebGL2() returns false, the hook returns a null ref and shouldRenderShader must evaluate to false. The existing card-shine CSS handles the fallback. Verify this works by temporarily forcing the function to return false.

Step 3: Visual calibration

Effects are seasoning (skills/shared-patterns/ui-design-judgment.md section 7): card name, cost, and rules text keep 4.5:1 contrast under the effect at peak intensity. Screenshot each tier at lg and xl with whatever renderer the harness has and look before tuning.

  • Legendary: Rainbow visible but not garish. Effective opacity 0.65–0.75 over the card. Animation speed: u_time advances at 0.5× real-time (not 1:1 — too fast feels cheap).
  • Rare: Blue hue shift + shimmer. Feels premium, not like a cursor glow effect.
  • Uncommon: Barely perceptible silver shimmer. If you notice it immediately on a static card, the opacity is too high.
  • Mouse tilt: The holographic angle shift should feel like physically tilting a card under light, not like a flashlight following the cursor. Limit the angular response to ±15 degrees of apparent rotation.

Step 4: Test across all rendered sizes

SizeWidthShader?Note
xs80pxNoToo small — no overhead
sm110pxNoToo small — no overhead
md140pxOptionalTest legibility first
lg170pxYesMinimum size for full effect
xl200pxYesPrimary target — should look best

Gate: All DevTools performance targets met. Mobile fallback verified. Visual quality approved at lg and xl sizes across all three shader tiers.


Error Handling

"WebGL: INVALID_OPERATION: useProgram: program not valid"

Shader compilation failed silently. Call gl.getShaderInfoLog(shader) immediately after gl.compileShader(shader). Common causes: GLSL syntax error, wrong #version 300 es directive missing, or a uniform declared but never referenced (GLSL compilers strip unused uniforms — reference them or remove the declaration).

Canvas present but effect invisible

Check mixBlendMode. On very dark card backgrounds, screen blend mode makes dark shader output invisible. For debugging, switch to normal blend mode to see the raw shader output. Also verify the canvas zIndex is above the frame PNG (15 > 10).

Browser console: "Too many active WebGL contexts"

The shared context singleton in useCardShader is not being used — individual hook calls are each creating a new context. Verify the module-level singleton is initialized once and reused. See the singleton pattern in references/shader-integration-react.md.

Canvas appears stretched or distorted on high-DPI displays

Canvas width / height attributes must match physical pixel dimensions, not CSS dimensions. CSS w-full h-full sets display size only. Use a ResizeObserver on the canvas element: canvas.width = entry.contentRect.width * devicePixelRatio.

TypeScript error: "Property 'ref' does not exist on 'HTMLCanvasElement'"

React 19 passes ref as a prop. The canvas element should be <canvas ref={shaderCanvasRef} ... /> — no forwardRef needed. Ensure the ref type matches: useRef<HTMLCanvasElement>(null).


Deep References

Load the reference needed for the current phase. All three total ~1,400 lines -- load all only if implementing everything in one pass.

SignalReference
GLSL fragment shader source for each tierreferences/card-shader-patterns.md
React 19 WebGL hook, context singletonreferences/shader-integration-react.md
Balatro holographic foil GLSL breakdownreferences/balatro-shader-breakdown.md

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Webgl Card Effects AI skill do?

Standalone WebGL fragment shaders for card visual effects: holographic foil, shimmer, rarity glow.

Why use Webgl Card Effects on TypingMind?

Because you install it once and use it with any model. Webgl Card Effects is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Webgl Card Effects in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/notque/vexjoy-agent/tree/main/skills/frontend/webgl-card-effects. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Webgl Card Effects?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Webgl Card Effects?

As many as you like. As long as a model supports skills, you can use Webgl Card Effects with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Webgl Card Effects AI skill free?

Yes. It is published on GitHub by notque under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇