So far, the Snippets only printed values.

Real Node, with your own access to your computer.

A Node Snippet runs on real Node, with your access to your files, the network and other programs. Code that reaches them waits for you to press Run.

Write a real file, after you press Run

Code that reaches your files, the network or other programs doesn’t run as you type: a shield in the rail says why. Press Run, and the file is written to the Working Folder and read back.

import { readFile, stat, writeFile } from "node:fs/promises"

1⌄import { readFile, stat, writeFile } from "node:fs/promises"

2⌄

3⌄const order = { pizza: "margherita", size: "large", extraCheese: true }

4⌄await writeFile("order.json", JSON.stringify(order, null, 2))

5⌄

6⌄const saved = JSON.parse(await readFile("order.json", "utf8"))

7⌄console.log(saved)

8⌄console.log((await stat("order.json")).size, "bytes on disk")

Object

pizza : margherita

size : large

extraCheese : true

69 bytes on disk

Ready·Runs on demand

The shield says why it waited. After Run, order.json is on disk, 69 bytes, and read back.

Call an API, start a server

fetch and node:http work as in Node. A server keeps listening until you press Stop.

import { createServer } from "node:http"

1⌄import { createServer } from "node:http"

2⌄

3⌄const server = createServer((request, response) => {

4⌄ console.log(request.method, request.url)

5⌄ response.setHeader("Content-Type", "application/json")

6⌄ response.end(JSON.stringify({ pizza: "on its way" }))

7⌄})

8⌄server.listen(3000, () => console.log("Listening on :3000"))

9⌄

10⌄const reply = await fetch("http://localhost:3000/orders/42")

11⌄console.log(await reply.json())

Listening on :3000

GET /orders/42

Object

L11

pizza : on its way

Ready·Runs on demand

A real server on port 3000, and a real request to it. It keeps listening until Stop.

Run other programs

child_process runs a program on your Mac and reads its output.

import { execSync } from "node:child_process"

1⌄import { execSync } from "node:child_process"

2⌄

3⌄const run = (cmd) => execSync(cmd, { encoding: "utf8" }).trim()

4⌄console.log(run("sw_vers -productName"), "on", run("uname -m"))

macOS on arm64

Ready·Runs on demand

sw_vers and uname run on your Mac, and their output reads back as macOS on arm64.

One Working Folder

Every Node Run starts in the Working Folder. Reveal it in Finder, move it to a folder of your own, or empty it.

General

Editor

Console

Running

Timeline

Starters

License

About

JSparkVersion 1.3.0

License…Get JSpark+

Node snippets: Node 24.21.0
Web snippets: Chromium 152.0.7977.130
JavaScript engine: V8 15.2.124.28

JSpark is up to dateLast checked today at 9:41

Check for Updates

Automatic updates

Downloads new versions in the background and installs them when you quit.

Copy DiagnosticsVersions, edition and macOS, for a problem report. Never your code or files.

Legal

License Agreement
Privacy Policy
Refund Policy

Show AcknowledgementsThe open-source software JSpark is built on, with each project’s license.

Working folder

12 KB

/Users/you/Library/Application Support/jspark/Working Folder

Reveal in FinderChange…Empty…

Where Node snippets run: a file written to a relative path, or beside __dirname, lands here and stays until you empty it.

Restore Default Settings…Your tabs, the Library and installed packages stay as they are.

Settings → About: the Working Folder’s path and size, with Reveal in Finder, Change… and Empty….

An API key, out of your code

Keep up to 3 environment variables in Settings, and every Node Run reads them as process.env.

environment

Running

Environment variables

Passed to every Node Run as process.env, and to a page’s server. Kept on this Mac, encrypted with a key in your Keychain, and never in the Library or an export.

API_KEY••••••••
DATABASE_URL••••••••

NameAPI_KEY

Value

Add

JSpark+ also passes the Working Folder’s .env to Runs.

Two variables saved, their values shown as dots. In JSpark, you can keep up to 3.

Run in a folder of your own JSpark+

Run in Folder… runs one Snippet in your project, so it can import your files and use your packages.

import { slugify } from "./utils.js"

1⌄import { slugify } from "./utils.js"

2⌄

3⌄console.log(slugify("Hello from my project"))

hello-from-my-project

Done in 12ms·in my-project×·Runs on demand

./utils.js is a file in your project. The footer names the folder the Snippet runs in.

Every variable, and your .env JSpark+

Keep as many variables as you need, and the Working Folder’s .env is passed to every Node Run too.

environment

Running

Environment variables

Passed to every Node Run as process.env, and to a page’s server. Kept on this Mac, encrypted with a key in your Keychain, and never in the Library or an export.

API_KEY••••••••
DATABASE_URL••••••••
STRIPE_KEY••••••••
SENTRY_DSN••••••••

NameAPI_KEY

Value

Add

The Working Folder’s .env is passed to Runs too, and wins over these.

Four variables, and the Working Folder’s .env as well, which wins over these.

Good to know

Not a sandbox: JSpark doesn’t limit a Node Run: code that deletes a file deletes it. Waiting for Run only keeps half-written code from starting.
Reading process.env doesn’t wait: Reading it runs as you type; changing it waits for Run.
Where the values are kept: Variable values stay on this Mac, encrypted with a key in your Keychain, never in the Library, an export or a problem report.
Servers keep running: The time limit stops only code that is busy without a pause, so a server keeps running until you press Stop.

Test code that reads files or calls an API The Working Folder Environment variables All features

When the code needs Node, it’s already here.

Download free for Mac

Not a trial: it doesn’t expire.

Windows: coming later