Claude Code for UnJS Ecosystem (2026)

The Setup

You are building JavaScript applications with UnJS, a collection of universal JavaScript utilities that work across Node.js, Deno, Bun, Cloudflare Workers, and browsers. UnJS includes packages like ofetch (fetch), consola (logging), defu (object merging), pathe (paths), and many more. Claude Code can write JavaScript utilities, but it reaches for heavy npm packages or Node.js-specific APIs instead of UnJS’s lightweight, universal alternatives.

What Claude Code Gets Wrong By Default

  1. Uses axios for HTTP requests. Claude installs axios with 400KB+ bundle. UnJS ofetch is lighter, has automatic JSON parsing, retry support, and works on every runtime — Node.js, Deno, Workers, and browsers.

  2. Uses path from Node.js. Claude imports import path from 'node:path'. This breaks in non-Node runtimes. UnJS pathe provides the same API but works universally across all JavaScript runtimes.

  3. Writes custom utility functions. Claude writes deep merge, type checking, and string manipulation helpers from scratch. UnJS has defu (deep defaults), untyped (runtime type validation), and scule (string case conversion) — battle-tested and universal.

  4. Ignores runtime compatibility. Claude uses Node.js APIs like process.env, fs, and child_process directly. UnJS provides std-env for environment detection, unenv for polyfilling Node APIs, and unbuild for building universal packages.

The CLAUDE.md Configuration

# UnJS Ecosystem Project
## Utilities
- Ecosystem: UnJS (universal JavaScript utilities)
- Runtime: Node.js, Deno, Bun, Workers, browsers
- Philosophy: lightweight, zero-dependency, universal
## UnJS Packages
- ofetch: universal fetch with retry and interceptors
- consola: universal logging with levels and reporters
- defu: deep object defaults merging
- pathe: universal path utilities (replaces node:path)
- scule: string case conversion (camel, pascal, kebab)
- ufo: URL manipulation utilities
- hookable: async hook/plugin system
- unenv: environment-agnostic Node.js polyfills
- std-env: runtime/environment detection
- unbuild: universal build tool for packages
## Conventions
- Prefer UnJS over heavy npm alternatives
- ofetch over axios/node-fetch
- pathe over node:path
- consola over winston/pino for universal code
- defu for config merging with defaults
- Check runtime compatibility for target environments
- Use unbuild for building universal packages

Workflow Example

You want to create a universal configuration system that works across runtimes. Prompt Claude Code:

“Build a configuration loader using UnJS packages. Use defu for merging config defaults, consola for logging, ofetch for fetching remote config, pathe for resolving config file paths, and std-env to detect the runtime environment. It should work in Node.js, Deno, and Cloudflare Workers.”

Claude Code should import from UnJS packages, use std-env to detect the runtime, pathe to resolve config paths (falling back for non-filesystem runtimes), ofetch to load remote config, defu to merge user config over defaults, and consola for structured logging — all without any Node.js-specific APIs.

Common Pitfalls

  1. Mixing Node.js APIs with UnJS equivalents. Claude uses path.join() alongside pathe.join() in the same file. Pick one — mixing Node.js built-ins with UnJS equivalents creates inconsistency and breaks universal compatibility.

  2. Not checking runtime support. Claude uses ofetch features that may not work everywhere. While UnJS packages are universal, some features depend on runtime capabilities. Use std-env to detect the runtime and adapt behavior accordingly.

  3. Over-installing UnJS packages. Claude installs every UnJS package mentioned in documentation. Only install packages you actually use — UnJS packages are small individually but unnecessary dependencies add up.

Get started → Generate your project setup with our Project Starter.

Try it: Browse 155+ skills in our Skill Finder.

Common Questions

How do I get started with claude code for unjs ecosystem?

Begin with the setup instructions in this guide. Install the required dependencies, configure your environment, and test with a small project before scaling to your full codebase.

What are the prerequisites?

You need a working development environment with Node.js or Python installed. Familiarity with the command line and basic Git operations is helpful. No advanced AI knowledge is required.

Can I use this with my existing development workflow?

Yes. These techniques integrate with standard development tools and CI/CD pipelines. Start by adding them to a single project and expand once you have verified the benefits.

Where can I find more advanced techniques?

Explore the related resources below for deeper coverage. The Claude Code documentation and community forums also provide advanced patterns and real-world case studies.