Screenshot API

A JPEG of any page: the visible window or the whole page.

The Screenshot API opens a page in a real browser and returns a JPEG: the visible window, or the whole page from top to bottom.

Call it

curl https://data.upscrape.com/screenshot \  -H "Authorization: Bearer $UPSCRAPE_API_KEY" \  -H "Content-Type: application/json" \  -H "Prefer: wait=30" \  -d '{"url": "https://www.wikipedia.org/", "full_page": true}'

import base64import osimport requestsjob = requests.post(    "https://data.upscrape.com/screenshot",    headers={"Authorization": f"Bearer {os.environ['UPSCRAPE_API_KEY']}", "Prefer": "wait=30"},    json={"url": "https://www.wikipedia.org/", "full_page": True},).json()image = job["results"][0]["data"]["image"]with open("wikipedia.jpg", "wb") as f:    f.write(base64.b64decode(image["data"]))

import { writeFile } from "node:fs/promises";const res = await fetch("https://data.upscrape.com/screenshot", {  method: "POST",  headers: {    Authorization: `Bearer ${process.env.UPSCRAPE_API_KEY}`,    "Content-Type": "application/json",    Prefer: "wait=30",  },  body: JSON.stringify({ url: "https://www.wikipedia.org/", full_page: true }),});const job = await res.json();const image = job.results[0].data.image;await writeFile("wikipedia.jpg", Buffer.from(image.data, "base64"));

The image is in results[0].data.image: base64 in data, with its content_type and size in bytes. It travels inside the JSON, so there is no link to expire.

Set up the shot

Input	What it does
`full_page`	The whole page, top to bottom, instead of the visible window.
`viewport_width`, `viewport_height`	The browser window, 1440 by 900 by default; from 320 by 200 up to 3840 by 2160. Use 390 by 844 for a phone.
`wait_for_selector`	Wait until an element matching this CSS selector appears, for pages that fill in after they load.
`wait_ms`	Wait this many milliseconds before the shot, up to 10 seconds.

The image is a JPEG.

Something missing or unclear? Ask the team that builds the endpoints.