transformers-js

Build and validate local browser or Bun inference with Hugging Face Transformers.js, WebGPU or WASM backends, quantization, caching, and cleanup.

Skill by metaspartan

IndexedMIT★ 32

Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/metaspartan-transformers-js

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

About

Build and validate local browser or Bun inference with Hugging Face Transformers.js, WebGPU or WASM backends, quantization, caching, and cleanup.

Use cases: ml

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.

Tools: No tools found
Secrets: None asked for
Install scripts: None found

Network hosts it talks to 1

codeload.github.com · found in install command

Files it touches

~/.claude/skills/transformers-js

What's inside

Skill name: transformers-js
Description: Build and validate local browser or Bun inference with Hugging Face Transformers.js, WebGPU or WASM backends, quantization, caching, and cleanup.
License: MIT
Instructions: 36 lines in SKILL.md · view on GitHub

Files (1)

SKILL.md

SKILL.md (first lines)

# Transformers.js

Use `@huggingface/transformers` for JavaScript or TypeScript inference in Bun, Node-compatible runtimes, and browsers.

## Setup

```bash
bun add @huggingface/transformers
```

```ts
import { pipeline } from "@huggingface/transformers";

const classifier = await pipeline("text-classification");
try {
  const result = await classifier("Cybara runs this model locally.");
  console.log(result);
} finally {
  await classifier.dispose();
}
```

## Workflow

1. Choose a model whose card lists Transformers.js or ONNX compatibility for the required task.
2. Pin the model ID and revision when reproducibility matters.
3. Start with CPU/WASM compatibility, then enable WebGPU only after feature detection.
4. Select dtype or quantization from measured memory, latency, and quality requirements.
5. Configure the cache explicitly for desktop or server runtimes.
6. Warm the model before benchmarking and separate download time from inference latency.
7. Dispose pipelines and tensors when the workflow ends.
8. Test unavailable WebGPU, interrupted downloads, offline cache reuse, and malformed inputs.

Do not load untrusted custom model code. For browser applications, keep model downloads visible to the user and avoid blocking the main thread during initialization or long inference.

From metaspartan/cybara, MIT.

Signals

Stars: 32
Forks: 9
Open issues: 0
Last commit (this folder): Jul 30, 2026 (3 months ago)
Last release: Oct 24, 2026 (4 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License: MIT
Repository: github.com/metaspartan/cybara
Website: cybara.ai
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: GitHub github-graphql

More from metaspartan

Cite this page

transformers-js skill. Agent Mart. https://agentmart.dev/skills/metaspartan-transformers-js (last updated 2026-10-24)