Node.js · Code recipe
Take a screenshot of a website in Node.js
Save a full-page PNG of any website, in Node.js 18+ with the built-in fetch. Every program on this page runs as it stands — each one was run against a stub of the API before it was published.
By Roger Campos · Last updated: September 2026
TL;DR
To take a screenshot of a website in Node.js, POST its URL to https://urlpipe.dev/screenshot with sync set to true, decode the Base64 body with Buffer.from(text, "base64") and write the bytes to a .png file. It is a full-page capture from real Chrome for 1 credit; leave sync out and the result goes to your webhook instead.
Free plan, no credit card. 1,000 credits a month.
One POST to /screenshot loads the page in real Chrome, scrolls it top to bottom so lazy images load, and captures the whole document — not just the fold. The image comes back Base64-encoded in a text/plain body, so the Node.js work is two lines: decode it with Buffer.from(text, "base64") and write the bytes.
The request below also sets page_options.block_cookie_banners, because a consent dialog over the page is the most common reason a screenshot is useless. Every other knob — viewport, device scale, JPEG or WebP, one element by selector, dark mode — goes in screenshot_options.
Setup
Before you start
Nothing to install: fetch is global from Node 18. The files end in .mjs so Node reads them as ES modules and top-level await works without a wrapper function.
node --version # v18 or later
export URLPIPE_API_KEY="your_api_key"
The request
Save a screenshot as a PNG file
sync: true keeps the request open until the result is ready. fetch resolves on any status, 4xx and 5xx included, so check res.ok yourself. The body is Base64 text; Buffer.from(text, "base64") turns it back into bytes.
import { writeFile } from "node:fs/promises";
const res = await fetch("https://urlpipe.dev/screenshot", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.URLPIPE_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({ url: "https://example.com", page_options: { block_cookie_banners: true }, sync: true }),
// fetch has no timeout of its own; a sync call can take up to 60 s.
signal: AbortSignal.timeout(90_000),
});
if (!res.ok) throw new Error(`URLpipe answered ${res.status}: ${await res.text()}`);
const png = Buffer.from(await res.text(), "base64");
await writeFile("screenshot.png", png);
console.log(`Saved screenshot.png (${png.length} bytes)`);
Run it: node screenshot.mjs
Async
The async variant: a token, a webhook and a poll
Leave out sync and the answer is a token, straight away. setTimeout from node:timers/promises is the awaitable sleep, so the polling loop reads top to bottom.
import { setTimeout as sleep } from "node:timers/promises";
import { writeFile } from "node:fs/promises";
const API = "https://urlpipe.dev";
const headers = {
Authorization: `Bearer ${process.env.URLPIPE_API_KEY}`,
"Content-Type": "application/json",
};
// No sync: the request is accepted at once and the work carries on without you.
const accepted = await fetch(`${API}/screenshot`, {
method: "POST",
headers,
body: JSON.stringify({
url: "https://example.com",
page_options: { block_cookie_banners: true },
report_to: "https://your-app.com/webhooks/urlpipe",
labels: { customer: "acme" },
}),
});
if (!accepted.ok) throw new Error(`URLpipe answered ${accepted.status}: ${await accepted.text()}`);
const { token } = await accepted.json();
console.log(`Accepted ${token}`);
// The result is POSTed to report_to when it is ready. Polling by token is the
// other way to collect it: no endpoint needed, and a backup for the webhook.
let res;
for (let attempt = 0; attempt < 60; attempt++) {
res = await fetch(`${API}/result/${token}`, { headers });
if (res.status !== 202) break; // 202 means still processing
await sleep(2000);
}
if (res.status === 202) throw new Error("Still processing after two minutes; try the token again later.");
if (res.status === 422) throw new Error(`The analysis failed: ${(await res.json()).error}`);
if (res.status === 410) throw new Error("The result is past the 30-day window; send the request again.");
if (!res.ok) throw new Error(`URLpipe answered ${res.status}: ${await res.text()}`);
const png = Buffer.from(await res.text(), "base64");
await writeFile("screenshot.png", png);
console.log(`Saved screenshot.png (${png.length} bytes)`);
Run it: node screenshot_async.mjs
Errors
Handle errors and retries
Read the status before the body: a 401 answers in plain text, so res.json() would throw. .catch(() => ({})) turns any body that is not JSON into an empty object, and the status still says what happened.
import { setTimeout as sleep } from "node:timers/promises";
import { writeFile } from "node:fs/promises";
class URLpipeError extends Error {}
// POST a sync request and return the response, or throw URLpipeError.
async function urlpipe(path, payload, attempts = 5) {
for (let attempt = 0; attempt < attempts; attempt++) {
const res = await fetch(`https://urlpipe.dev${path}`, {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.URLPIPE_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({ ...payload, sync: true }),
signal: AbortSignal.timeout(90_000),
});
if (res.ok) return res;
if (res.status === 401) {
throw new URLpipeError("401: the API key is missing or wrong. Check URLPIPE_API_KEY.");
}
const body = await res.json().catch(() => ({}));
const code = body.error ?? "";
const detail = body.message ? `${code}: ${body.message}` : code;
if (res.status === 429 && code === "rate_limited") {
// Sending too fast: Retry-After says how long the window has left.
await sleep(Number(res.headers.get("Retry-After") ?? 1) * 1000);
} else if (res.status === 429 && code === "concurrency_limit") {
// Every parallel slot on your plan is busy with your own requests.
await sleep(2 ** attempt * 1000);
} else if (res.status === 504) {
// Still running on our side; the token collects it from GET /result/:token.
throw new URLpipeError(`504 processing_timeout: collect it later with token ${body.token}`);
} else {
// 403 email_unverified, 422 (a bad parameter, or a page that would not load),
// 429 quota_exceeded: sending the same request again gets the same answer.
throw new URLpipeError(`${res.status}: ${detail}`);
}
}
throw new URLpipeError(`429: still refused after ${attempts} attempts`);
}
let res;
try {
res = await urlpipe("/screenshot", { url: "https://example.com", page_options: { block_cookie_banners: true } });
} catch (error) {
console.error(`URLpipe: ${error.message}`);
process.exit(1);
}
const png = Buffer.from(await res.text(), "base64");
await writeFile("screenshot.png", png);
console.log(`Saved screenshot.png (${png.length} bytes)`);
Run it: node screenshot_errors.mjs
Details
What to know about /screenshot
- The default viewport is 1350 × 797 and the capture follows the page down to 16,384 px; a taller page is cut there.
- Every response carries an
X-Result-Urlheader: a link to the same image that needs no API key and stays valid for 30 days. Often you can store that link instead of the file. - PNG is the default;
"format": "jpeg"or"webp"inscreenshot_optionsmakes a long page far smaller. - It captures images, not PDFs, and it does not record video.
- A screenshot of the same URL with the same options is served from the cache for 7 days by default, free. Set
max_ageto refresh sooner.
Client library
Or use the official JavaScript client
The same call in one line, with retries that never bill twice, typed errors, long analyses waited out for you and webhook verification. TypeScript types, no dependencies; Node 18+, Bun, Deno and Cloudflare Workers.
// npm install @urlpipe/sdk
import Urlpipe from "@urlpipe/sdk";
const client = new Urlpipe(); // reads URLPIPE_API_KEY
const { data } = await client.markdown("https://example.com");
console.log(data);FAQ
Frequently asked questions
Why is the screenshot response Base64 and not an image?
How do I take a mobile screenshot?
Can I screenshot one element instead of the whole page?
Do I need an SDK to call URLpipe from Node.js?
Get a key and run it.
Free plan, no card. Paste your key into URLPIPE_API_KEY and every program on this page runs as it is.