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

@llmtrim/js

v0.3.2

Published

WebAssembly/JS bindings for the llmtrim-core compression engine (wasm-bindgen).

Readme

@llmtrim/js

WebAssembly/JS bindings for the llmtrim-core compression engine, for running the static prompt/payload compressor in a browser, Node, Bun, Deno, or a Cloudflare Worker. No network or filesystem access.

Published to npm as @llmtrim/js (with @llmtrim/wasm as an alias that re-exports the same package).

API

import { compress } from "@llmtrim/js";

const out = compress(requestBodyJson, "openai", "agent");
// out: CompressOutput { request_json, provider, model, tokenizer_label,
//   tokenizer_exact, input_tokens_before, input_tokens_after,
//   frozen_input_tokens, output_shaped, stages: StageReport[] }
  • provider: "openai" | "anthropic" | "google", or undefined/null to auto-detect from the body.
  • preset: a named workload preset (aggressive, agent, code, rag, safe, …), or undefined/null for the built-in defaults. This binding never reads the environment or a config file.

TypeScript types for CompressOutput and StageReport are generated via tsify, so the .d.ts is fully typed (no any).

This build links llmtrim-core with default-features = false: the estimate tokenizer is used (counts are approximate; savings percentages are unchanged), and code skeletonization and image downscaling are no-ops. That keeps the bundle small (~1 MB gzipped, under the Cloudflare Workers 3 MB free-tier cap).

Building

The npm package is built in CI with wasm-pack build --target bundler (it manages a matching wasm-bindgen internally). The manual cargo + wasm-bindgen recipe below is the equivalent for local development:

rustup target add wasm32-unknown-unknown
cargo install wasm-bindgen-cli

# The JS-backed getrandom backend needs a rustc cfg in the environment (it cannot live in a
# repo .cargo/config.toml, which would break `cargo publish`).
RUSTFLAGS='--cfg getrandom_backend="wasm_js"' \
  cargo build -p llmtrim-wasm --release --target wasm32-unknown-unknown

wasm-bindgen target/wasm32-unknown-unknown/release/llmtrim_wasm.wasm \
  --out-dir pkg --target bundler   # or: nodejs | web

# optional size pass
wasm-opt -O3 --enable-reference-types --enable-bulk-memory --enable-mutable-globals \
  --enable-nontrapping-float-to-int --enable-sign-ext --enable-multivalue \
  pkg/llmtrim_wasm_bg.wasm -o pkg/llmtrim_wasm_bg.wasm

Smoke test

After building with --target nodejs into pkg/, run node smoke.mjs (it imports ./pkg/llmtrim_wasm.js). It exercises OpenAI and Anthropic requests, a named preset, an error path, and non-ASCII (CJK) input.

License

MPL-2.0.