Pixijs Application logo

Pixijs Application

Organization
pixijs
pixijs-application

Use this skill when creating and configuring a PixiJS v8 Application. Covers new Application() + async app.init() options (width, height, background, antialias, resolution, autoDensity, preference, resizeTo, autoStart, sharedTicker, canvas, useBackBuffer, powerPreference, eventFeatures, accessibilityOptions, gcActive, bezierSmoothness, webgl/webgpu/canvasOptions per-renderer overrides), app.stage/renderer/canvas/screen/domContainerRoot access, ResizePlugin, TickerPlugin, CullerPlugin (cullable, cullArea), custom ApplicationPlugin via ExtensionType.Application, start/stop lifecycle, and app.destroy() with releaseGlobalResources. Triggers on: Application, app.init, app.stage, app.renderer, app.canvas, app.screen, app.domContainerRoot, ApplicationOptions, ApplicationPlugin, ExtensionType.Application, resizeTo, sharedTicker, skipExtensionImports, preferWebGLVersion, preserveDrawingBuffer, cullable, CullerPlugin, app.start, app.stop, app.destroy, releaseGlobalResources.

Overview

Publisherpixijs
Repositorypixijs-skills
Skill namepixijs-application
Stars
357
Forks
19
Bundled files
1
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.

  • 1 bundled files

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

  • Open source

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

Installation

Install the Pixijs Application 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/pixijs/pixijs-skills/tree/main/skills/pixijs-application
  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/pixijs/pixijs-skills.git /tmp/pixijs-skills
mkdir -p .claude/skills
cp -r /tmp/pixijs-skills/skills/pixijs-application .claude/skills/pixijs-application
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Pixijs Application 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 Pixijs Application 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 Pixijs Application 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.

Application is the convenience wrapper that owns a renderer, a root stage Container, a canvas, and the Ticker/Resize plugins. In v8 the constructor takes no arguments; all configuration is passed to the async app.init() call which instantiates the renderer via autoDetectRenderer.

Quick Start

ts
import { Application } from "pixi.js";

const app = new Application();

await app.init({
  resizeTo: window,
  background: "#1099bb",
  antialias: true,
  preference: "webgl",
  autoDensity: true,
  resolution: window.devicePixelRatio,
});

document.body.appendChild(app.canvas);

Related skills: pixijs-core-concepts (renderers, render pipeline), pixijs-ticker (render loop detail), pixijs-scene-container (working with app.stage), pixijs-environments (non-browser setups).

Core Patterns

Lifecycle: construct, init, render, destroy

ts
import { Application } from "pixi.js";

const app = new Application();

await app.init({ width: 800, height: 600 });
document.body.appendChild(app.canvas);

// ... run scene, ticker drives app.render() automatically ...

app.destroy(
  { removeView: true, releaseGlobalResources: true },
  { children: true, texture: true, textureSource: true },
);
  • new Application() allocates the instance but creates nothing. Options passed here are ignored with a v8 deprecation warning.
  • app.init(options) is async. It builds the renderer, wires up plugins, and must complete before you can use app.canvas, app.renderer, or app.screen.
  • The TickerPlugin calls app.render() every frame once init resolves (unless autoStart: false).
  • app.destroy(rendererDestroyOptions, stageDestroyOptions) — the first argument forwards to renderer.destroy(). Pass true or { removeView: true } to remove the canvas from the DOM. Add releaseGlobalResources: true to drain global pools (batches, texture caches) when tearing down and re-creating an app in the same tab; omitting it is the usual cause of flickering and stale textures after a re-init (see pixijs-performance).

Key init options

ts
await app.init({
  width: 800,
  height: 600,
  background: 0x1099bb,
  backgroundAlpha: 1,

  antialias: true,
  resolution: window.devicePixelRatio,
  autoDensity: true,

  preference: "webgpu",

  autoStart: true,
  sharedTicker: false,

  resizeTo: window,

  canvas: document.querySelector("#game-canvas") as HTMLCanvasElement,
});

For every option — view/canvas, background, renderer preference (including the array form), ticker, resize, culler, events, accessibility, WebGL/WebGPU context flags, Graphics bezier smoothness, GC, and per-renderer overrides (webgl / webgpu / canvasOptions) — see references/application-options.md.

Application properties

ts
app.stage; // root Container; add all display objects here
app.renderer; // the WebGL/WebGPU/Canvas renderer instance
app.canvas; // the HTMLCanvasElement (insert it into the DOM yourself)
app.screen; // Rectangle describing the visible area in CSS pixels
app.domContainerRoot; // HTMLDivElement that holds DOMContainer overlays

app.stage is a plain Container. For scene graph detail (transforms, addChild, destroy) see pixijs-scene-container. For renderer-level operations (extract, generateTexture, custom systems) see pixijs-core-concepts and pixijs-custom-rendering. app.domContainerRoot is the <div> that the renderer uses to host DOMContainer overlays; append it next to app.canvas when you need DOM elements pinned to scene nodes (see pixijs-scene-dom-container).

ResizePlugin

Set resizeTo at init (or reassign app.resizeTo later) to have the plugin listen for the resize event and call renderer.resize() with the target element's client size. Combine with autoDensity: true and resolution: window.devicePixelRatio for high-DPI output.

ts
await app.init({ resizeTo: window });

app.resizeTo = document.querySelector("#game-container") as HTMLElement;

app.resize(); // immediate resize to the target's current size
app.queueResize(); // defer the resize to the next animation frame
app.cancelResize(); // drop a pending queueResize

The plugin keeps the canvas matched to the target. app.screen and app.canvas.width/height update in response; read them after the resize to place UI.

  • app.resize() — immediate synchronous resize.
  • app.queueResize() — coalesces rapid calls by deferring to the next frame; internally used by the window.resize listener to avoid redundant work.
  • app.cancelResize() — cancels a queued resize. Call this before tearing down your own layout code that triggered queueResize.

Ticker basics

The TickerPlugin creates app.ticker and registers app.render() on it at UPDATE_PRIORITY.LOW. Control the loop with app.start()/app.stop() and add callbacks with app.ticker.add / app.ticker.addOnce:

ts
app.ticker.add((ticker) => {
  sprite.rotation += 0.01 * ticker.deltaTime;
});

app.ticker.addOnce(() => {
  console.log("runs once on the next frame, then removes itself");
});

app.stop(); // pause the render loop (e.g. tab hidden)
app.start(); // resume

The callback receives the Ticker instance; read ticker.deltaTime for a frame-rate-independent multiplier (~1.0 at 60fps), ticker.deltaMS for real milliseconds, or ticker.FPS for the current frame rate. See pixijs-ticker for priorities, FPS capping, onRender, shared vs private tickers, and the v8 callback signature change.

Manual render loop

ts
await app.init({ autoStart: false, width: 800, height: 600 });
document.body.appendChild(app.canvas);

function frame() {
  updateScene();
  app.render();
  requestAnimationFrame(frame);
}
frame();

autoStart: false prevents the TickerPlugin from starting the ticker automatically. Call app.render() yourself (or app.renderer.render({ container: app.stage }) for the same effect). If you still want registered ticker callbacks to fire, call app.ticker.update() inside your loop before app.render().

CullerPlugin (opt-in)

The CullerPlugin skips rendering containers that fall outside app.renderer.screen. It isn't registered by default; add it before creating your app:

ts
import {
  Application,
  Container,
  Sprite,
  extensions,
  CullerPlugin,
  Rectangle,
} from "pixi.js";

extensions.add(CullerPlugin);

const app = new Application();
await app.init({ width: 800, height: 600 });

const world = new Container();
world.cullable = true; // this container is culled when its bounds leave the screen
world.cullableChildren = true; // default; set `false` to skip recursing into children

const tile = Sprite.from("tile.png");
tile.cullable = true;
world.addChild(tile);
app.stage.addChild(world);

Containers are not culled unless cullable is set. Override the default bounds check with container.cullArea = new Rectangle(x, y, w, h) when child bounds are expensive to compute. The plugin wraps app.render() so Culler.shared.cull(app.stage, app.renderer.screen) runs before every frame. See pixijs-performance for when culling pays off.

Custom Application plugins

Extend Application by registering a class with static init, static destroy, and static extension = ExtensionType.Application. Both methods are called with this bound to the Application instance, so this.renderer and this.stage are available.

ts
import {
  Application,
  ExtensionType,
  extensions,
  type ApplicationOptions,
} from "pixi.js";

class FpsOverlay {
  public static extension = ExtensionType.Application;

  public static init(this: Application, options: Partial<ApplicationOptions>) {
    // runs inside app.init() after the renderer is created
    // attach props/methods to `this` to expose them on the app
  }

  public static destroy(this: Application) {
    // runs inside app.destroy() — tear down anything you attached
  }
}

extensions.add(FpsOverlay);

Plugins initialize in registration order and destroy in reverse. To add typed options for your plugin, extend PixiMixins.ApplicationOptions:

ts
declare global {
  namespace PixiMixins {
    interface ApplicationOptions {
      fpsOverlay?: { visible?: boolean };
    }
  }
}

await app.init({ fpsOverlay: { visible: true } });

The built-in ResizePlugin, TickerPlugin, and opt-in CullerPlugin all use this same contract. If you set skipExtensionImports: true, register the built-ins you need yourself (extensions.add(ResizePlugin, TickerPlugin)).

Common Mistakes

[CRITICAL] Passing options to the constructor

Wrong:

ts
const app = new Application({ width: 800, height: 600 });
document.body.appendChild(app.canvas);

Correct:

ts
const app = new Application();
await app.init({ width: 800, height: 600 });
document.body.appendChild(app.canvas);

In v8 the Application constructor takes no arguments. Options passed there are ignored and log a deprecation warning; the renderer is only created inside the async init() call.

[HIGH] Using app.view instead of app.canvas

Wrong:

ts
document.body.appendChild(app.view);

Correct:

ts
document.body.appendChild(app.canvas);

app.view was renamed to app.canvas in v8. The old getter still works but emits a deprecation warning.

[MEDIUM] Touching app.canvas or app.renderer before init resolves

Wrong:

ts
const app = new Application();
document.body.appendChild(app.canvas);
app.init({ width: 800, height: 600 });

Correct:

ts
const app = new Application();
await app.init({ width: 800, height: 600 });
document.body.appendChild(app.canvas);

app.renderer, app.canvas, and app.screen are only populated once the init() promise resolves. Accessing them earlier returns undefined.

API Reference

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 Pixijs Application AI skill do?

Use this skill when creating and configuring a PixiJS v8 Application. Covers new Application() + async app.init() options (width, height, background, antialias, resolution, autoDensity, preference, resizeTo, autoStart, sharedTicker, canvas, useBackBuffer, powerPreference, eventFeatures, accessibilityOptions, gcActive, bezierSmoothness, webgl/webgpu/canvasOptions per-renderer overrides), app.stage/renderer/canvas/screen/domContainerRoot access, ResizePlugin, TickerPlugin, CullerPlugin (cullable, cullArea), custom ApplicationPlugin via ExtensionType.Application, start/stop lifecycle, and app....

Why use Pixijs Application on TypingMind?

Because you install it once and use it with any model. Pixijs Application 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 Pixijs Application in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/pixijs/pixijs-skills/tree/main/skills/pixijs-application. 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 Pixijs Application?

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 Pixijs Application?

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

Is the Pixijs Application AI skill free?

Yes. It is published on GitHub by pixijs 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 👇