npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@desert-ant-labs/shapes

v3.2.0

Published

On-device single-stroke shape recognition for JavaScript. Runs in the browser (WebAssembly + LiteRT.js) and server-side in Node (native), from one import.

Readme

@desert-ant-labs/shapes

On-device single-stroke shape recognition for JavaScript that runs in the browser and in Node. Give Shapes one hand-drawn stroke and it returns a clean vector shape: a line, rectangle, triangle, ellipse, or star. Strokes stay local.

A small classifier proposes a shape, a geometric fitter produces the clean parameters, and the stroke is accepted only if it clears that class's calibrated confidence and residual gates. The result snaps to nice axes, circles, squares, and 15 degree rotations.

Two entries share one Shapes API:

  • @desert-ant-labs/shapes (default): a WebAssembly pipeline with LiteRT.js inference, for the browser. It has no native dependencies, so a single import builds cleanly for every target of a multi-target bundler (Next.js, Remix, SvelteKit, Nuxt), including the browser bundle and the Client-Component SSR pass those frameworks render in Node. It is safe to import during server-side rendering, but LiteRT.js needs a browser (or Web Worker) to initialize, so Shapes.load() runs inference only in the browser; calling it in plain Node throws an actionable error pointing you to /native.
  • @desert-ant-labs/shapes/native: a prebuilt native core, Core ML on macOS and LiteRT on Linux, for server-side inference in Node. No @litertjs/core, no build tools, no flags. Import it from server-only code (API routes, server actions, plain Node scripts). Do not import it from a component that also renders in the browser.
# Browser (default entry):
npm i @desert-ant-labs/shapes @litertjs/core

# Server-side inference in Node (/native entry) needs no extra install:
npm i @desert-ant-labs/shapes

The model is downloaded from the Hugging Face Hub on first use at the SDK's pinned tag, then cached. Nothing model-sized is shipped in the npm tarball.

import { Shapes } from "@desert-ant-labs/shapes";

const shapes = await Shapes.load();
const shape = await shapes.recognize(points);   // [{x, y}, ...] or [x0, y0, ...]
// { kind: "ellipse", center: {x, y}, semiMajor, semiMinor, rotation }

shapes.dispose(); // release the model (both builds)

Server-only code that wants the native core imports the same API from the /native subpath:

import { Shapes } from "@desert-ant-labs/shapes/native"; // server only

Shapes

recognize returns null when the stroke is rejected or degenerate, otherwise one of these, discriminated by kind:

  • line - from, to
  • rectangle - corners, four points around the perimeter
  • triangle - vertices, three points
  • ellipse - center, semiMajor, semiMinor, rotation (radians)
  • star - center, outerRadius, innerRadius, rotation (radians), pointCount

The kinds and field names are identical in Swift and Kotlin, so a stroke recognized on one platform describes itself the same way on every other.

recognize(points, { minimumConfidence }) raises the classifier threshold on top of each class's calibrated gate; the default 0 applies only the model's own gates.

Loading the model

By default Shapes.load() downloads the model files from the Hugging Face Hub (desert-ant-labs/shapes) at the SDK's pinned tag, verifies them, and caches them. The browser build fetches the .tflite for LiteRT.js and caches it in the browser. The native build (/native) fetches the .tflite on Linux or the .mlmodelc/ on macOS and caches it under the OS cache dir.

To self-host or run fully offline, opt out of the Hub:

  • directory: an explicit model directory (native build, or the browser build under Node). Files already there are used offline, otherwise the model is downloaded into it.
  • modelBaseUrl: a base URL you serve the model files from, for example "/assets/shapes/" (browser build).

Shapes.load() also accepts:

  • cacheRoot: base directory for the managed on-disk cache, default ~/.cache (native build, or the browser build under Node).
  • onProgress: load or download progress callback, fraction in [0, 1].

Browser-build-only options:

  • litert: a bring-your-own @litertjs/core module.
  • litertWasmDir: URL or path to the LiteRT.js Wasm directory.
  • accelerator: one of "wasm", "webgpu", or "webnn".

Bundlers and SSR

The default @desert-ant-labs/shapes import is safe to use directly in components: it is pure JavaScript + WebAssembly with no native modules, so bundlers can build it for the browser and for the Node SSR pass from the same module graph with no configuration.

The @desert-ant-labs/shapes/native subpath loads a native addon (via koffi) and is for server-only code. If you import it inside a framework that bundles server code (for example a Next.js Route Handler or Server Action), mark it external so the bundler does not try to bundle the native binary. In Next.js:

// next.config.js
module.exports = { serverExternalPackages: ["@desert-ant-labs/shapes"] };

Platforms

The native server build (/native) ships for linux-x64, linux-arm64, and darwin-arm64. Other Node platforms throw a clear error at load(); use the default WebAssembly build, the Swift package, or a browser for those.

License

Desert Ant Labs Source-Available License 1.0: free below 100,000 monthly active devices per platform. Above that a commercial license is required. Full terms: https://license.desertant.com/1.0