Network-tab-style DevTools for LLM calls in Angular apps. See every prompt, response, token, and dollar your app spends — without leaving the browser.

JWA, JWS, JWE, JWT, JWK, JWKS for Node.js, Browser, Cloudflare Workers, Deno, Bun, and other Web-interoperable runtimes

Whole-log conversation counts and wall times projection (sessionStats) for the DeepSeek Harness

Token-Oriented Object Notation (TOON) – compact, human-readable serialization of JSON data for LLM prompts

Auto-convert clipboard JSON to TOON — 40% fewer LLM tokens

Polyfill of future proposal for `util.parseArgs()`

Custom transforms for Style-Dictionary, to work with Design Tokens that are exported from Tokens Studio

Theme and context provider for the WordPress Design System.

A JavaScript text diff implementation.

 

Fast token estimation with 95%+ average accuracy in a 2kB bundle

Convert Puppeteer scripts to Tap Plan v2 — deterministic browser-automation plans for Claude Code, Cursor, and any MCP host. Zero LLM tokens at replay.

Design tokens are the single source of truth to name and store design decisions.

Generate string from a token array by interpolating values.

Convert Playwright scripts into Tap v2 plans. Compile .spec.ts / .test.ts files into deterministic v2 Plan JSON that runs at zero LLM tokens in Claude Code, Cursor, and any MCP host.

Zero-dependency OTLP span contract for agent traces: span shape, gen_ai attribute vocabulary, pure builders, and a conformance validator that reports what a trace can and cannot answer

Gotong workflow + agent prompt eval harness — lightweight structural checkers that catch output-format regressions without burning LLM tokens.

Display language model outputs in your React project.

Capix Code — decentralized AI coding agent with GPU marketplace

HTTP request logger middleware for node.js