Capture screenshots of websites
It uses Puppeteer (Chrome) under the hood.
See capture-website-cli for the command-line tool.
npm install capture-websiteNote to Linux users: If you get a sandbox-related error, you need to enable system sandboxing.
import captureWebsite from 'capture-website';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png');Capture a screenshot of the given input and save it to the given outputFilePath.
Intermediate directories are created for you if they do not exist.
Returns a Promise<void> that resolves when the screenshot is written.
Capture a screenshot of the given input.
Returns a Promise<Uint8Array> with the screenshot as binary.
Capture a screenshot of the given input.
Returns a Promise<string> with the screenshot as Base64.
Type: string
The URL, file URL, data URL, local file path to the website, or HTML.
import captureWebsite from 'capture-website';
await captureWebsite.file('index.html', 'local-file.png');Type: object
Type: string
Default: 'url'
Values: 'url' | 'html'
Set it to html to treat input as HTML content.
import captureWebsite from 'capture-website';
await captureWebsite.file('<h1>Awesome!</h1>', 'screenshot.png', {
inputType: 'html'
});Type: number
Default: 1280
Page width.
Type: number
Default: 800
Page height.
Type: string
Values: 'png' | 'jpeg' | 'webp' | 'pdf'
Default: 'png'
Output type.
Note
When using 'pdf', the clip, element, and quality options are not supported. Use the pdf option for PDF-specific settings.
Type: number
Values: 0...1
Default: 1
Image quality. Only for {type: 'jpeg'} and {type: 'webp'}.
Type: number
Default: 2
Scale the webpage n times.
The default is what you would get if you captured a normal screenshot on a computer with a retina (High DPI) screen.
Type: string
Values: Devices (Use the name property)
Make it look like the screenshot was taken on the specified device.
This overrides the width, height, scaleFactor, and userAgent options.
import captureWebsite from 'capture-website';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png', {
emulateDevice: 'iPhone X'
});Type: boolean
Default: false
Capture the full scrollable page, not just the viewport.
Warning
Screenshots taller than 16384px will repeat content due to a Chromium limitation. Use the clip option to capture sections separately.
Type: boolean
Default: false
Scroll through the entire page to trigger lazy-loaded content before capturing the screenshot.
This scrolls through the page viewport by viewport, waiting for network idle after each scroll, then returns to the original scroll position. This is useful for capturing third-party widgets, marketing banners, or any content that only loads when scrolled into view.
Note
This option is automatically enabled when fullPage is true.
import captureWebsite from 'capture-website';
await captureWebsite.file('https://example.com', 'screenshot.png', {
preloadLazyContent: true
});Type: boolean
Default: true
Include the default white background.
Disabling this lets you capture screenshots with transparency.
Type: number (seconds)
Default: 60
The number of seconds before giving up trying to load the page.
Specify 0 to disable the timeout.
Type: boolean
Default: false
Throw an error if the HTTP response status code is not in the 200-299 range.
This only applies to http:// and https:// URLs. Local files, data URLs, and HTML content are not affected.
import captureWebsite from 'capture-website';
// Throws an error if the page returns a 404, 500, etc.
await captureWebsite.file('https://example.com/missing', 'screenshot.png', {
throwOnHttpError: true
});Type: number (seconds)
Default: 0
The number of seconds to wait after the page finished loading before capturing the screenshot.
This can be useful if you know the page has animations that you like it to finish before capturing the screenshot.
This is also useful for capturing lazy-loaded content from third-party scripts (marketing widgets, analytics, etc.) that load asynchronously after the initial page load.
Type: string
Wait for a DOM element matching the given CSS selector to appear in the page and to be visible before capturing the screenshot. It times out after options.timeout seconds.
Type: string
Capture the DOM element matching the given CSS selector. It will wait for the element to appear in the page and to be visible. It times out after options.timeout seconds. Any actions performed as part of options.beforeScreenshot occur before this.
Type: string[]
Hide DOM elements matching the given CSS selectors.
Can be useful for cleaning up the page.
This sets visibility: hidden on the matched elements.
import captureWebsite from 'capture-website';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png', {
hideElements: [
'#sidebar',
'img.ad'
]
});Type: string[]
Remove DOM elements matching the given CSS selectors.
This sets display: none on the matched elements, so it could potentially break the website layout.
Type: string
Click the DOM element matching the given CSS selector.
Type: string | object
Scroll to the DOM element matching the given CSS selector.
Type: string
A CSS selector.
Type: string
Values: 'top' | 'right' | 'bottom' | 'left'
Offset origin.
Type: number
Offset in pixels.
Type: boolean
Default: false
Disable CSS animations and transitions.
Type: boolean
Default: true
Type: boolean
Default: true
Whether JavaScript on the website should be executed.
This does not affect the scripts and modules options.
Type: string[]
Inject JavaScript modules into the page.
Accepts an array of inline code, absolute URLs, and local file paths (must have a .js extension).
import captureWebsite from 'capture-website';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png', {
modules: [
'https://sindresorhus.com/remote-file.js',
'local-file.js',
`
document.body.style.backgroundColor = 'red';
`
]
});Type: string[]
Same as the modules option, but instead injects the code as <script> instead of <script type="module">. Prefer the modules option whenever possible.
Type: string[]
Inject CSS styles into the page.
Accepts an array of inline code, absolute URLs, and local file paths (must have a .css extension).
import captureWebsite from 'capture-website';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png', {
styles: [
'https://sindresorhus.com/remote-file.css',
'local-file.css',
`
body {
background-color: red;
}
`
]
});Type: object
Default: {}
Set custom HTTP headers.
import captureWebsite from 'capture-website';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png', {
headers: {
'x-powered-by': 'https://github.com/sindresorhus/capture-website'
}
});Type: string
Set a custom user agent.
Type: string
Set a custom referrer header for the page navigation.
This takes preference over the referrer header value set by the headers option.
import captureWebsite from 'capture-website';
await captureWebsite.file('https://example.com', 'screenshot.png', {
referrer: 'https://google.com'
});Type: object
PDF-specific options. Only applies when type is 'pdf'.
The object can have the following properties:
format(string): Paper format. Values:'letter' | 'legal' | 'tabloid' | 'ledger' | 'a0' | 'a1' | 'a2' | 'a3' | 'a4' | 'a5' | 'a6'. Default:'letter'landscape(boolean): Use landscape orientation. Default:falsemargin(object): Set page margins. Accepts string values (e.g., '10px', '2cm') or numeric values in pixels. Object withtop,right,bottom,leftproperties.background(boolean): Include background graphics. Default:false
import captureWebsite from 'capture-website';
await captureWebsite.file('https://example.com', 'output.pdf', {
type: 'pdf',
pdf: {
format: 'a4',
landscape: true,
margin: {
top: '20px',
right: '20px',
bottom: '20px',
left: '20px'
},
background: true
}
});Type: Array<string | object>
Set cookies in browser string format or object format.
Tip: Go to the website you want a cookie for and copy-paste it from DevTools.
import captureWebsite from 'capture-website';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png', {
cookies: [
// This format is useful for when you copy it from the browser
'id=unicorn; Expires=Wed, 21 Oct 2018 07:28:00 GMT;',
// This format is useful for when you have to manually create a cookie
{
name: 'id',
value: 'unicorn',
expires: Math.round(new Date('2018-10-21').getTime() / 1000)
}
]
});Type: object
Credentials for HTTP authentication.
Type: string
Type: string
Type: Function
The specified function is called right before navigating to the page. It receives the Puppeteer Page instance as the first argument and the browser instance as the second argument. This is useful for registering event handlers that need to be set up before page navigation, such as dialog handlers or request interceptors. The function can be async.
Caution
Do not register event listeners on the browser object. Browser listeners persist and cause memory leaks. Only register listeners on the page object.
Note: Do not call page.close(), browser.close(), or page.goto().
Example: Handling a dialog during page load
import captureWebsite from 'capture-website';
await captureWebsite.file('https://example.com', 'screenshot.png', {
beforeNavigation: async page => {
page.once('dialog', async dialog => {
await dialog.dismiss();
});
}
});Example: Blocking requests
import captureWebsite from 'capture-website';
await captureWebsite.file('https://example.com', 'screenshot.png', {
beforeNavigation: async page => {
await page.setRequestInterception(true);
// Important: Every request must call continue() or abort()
page.on('request', request => {
if (['image', 'font'].includes(request.resourceType())) {
request.abort();
} else {
request.continue();
}
});
}
});Type: Function
The specified function is called right before the screenshot is captured, as well as before any bounding rectangle is calculated as part of options.element. It receives the Puppeteer Page instance as the first argument and the browser instance as the second argument. This gives you a lot of power to do custom stuff. The function can be async.
Note: Make sure to not call page.close() or browser.close().
import captureWebsite from 'capture-website';
import checkSomething from './check-something.js';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png', {
beforeScreenshot: async (page, browser) => {
await checkSomething();
await page.click('#activate-button');
await page.waitForSelector('.finished');
}
});Type: boolean
Default: false
Show the browser window so you can see what it's doing, redirect page console output to the terminal, and slow down each Puppeteer operation.
Note: This overrides launchOptions with {headless: false, slowMo: 100}.
Type: Function
A function that is called whenever the page logs to its console. This allows you to receive and handle console output from the page.
The function receives a Puppeteer ConsoleMessage object as its argument, which provides access to the console message text, type, location, and more.
Note: Errors thrown in the callback are caught to prevent breaking the screenshot capture. If debug is enabled, errors are logged to the console.
import captureWebsite from 'capture-website';
await captureWebsite.file('https://example.com', 'screenshot.png', {
onConsole: message => {
console.log(`[${message.type()}] ${message.text()}`);
}
});Type: boolean
Default: false
Emulate preference of dark color scheme (prefers-color-scheme).
Type: object | number
Default: 0
Inset the bounding box of the screenshot.
Accepts an object {top?: number; right?: number; bottom?: number; left?: number} or a number as a shorthand for all directions.
Positive values, for example inset: 10, will decrease the size of the screenshot.
Negative values, for example inset: {left: -10}, will increase the size of the screenshot.
Note: This option is ignored if the fullPage option is set to true. Can be combined with the element option.
Note: When the width or height of the screenshot is equal to 0 an error is thrown.
Example: Include 10 pixels around the element.
import captureWebsite from 'capture-website';
await captureWebsite.file('index.html', 'screenshot.png', {
element: '.logo',
inset: -10
});Example: Ignore 15 pixels from the top of the viewport.
import captureWebsite from 'capture-website';
await captureWebsite.file('index.html', 'screenshot.png', {
inset: {
top: 15
}
});Type: boolean
Default: false
Allow cross-origin requests. Useful when capturing local HTML files that reference other local files.
Warning: This disables web security features and should only be used in trusted environments.
await captureWebsite.file('file:///path/to/local/file.html', 'screenshot.png', {
allowCORS: true
});Type: boolean
Default: false
Wait for all network connections to finish before capturing the screenshot.
This can be useful for websites that have long-running requests or many resources to load.
Note: This uses Puppeteer's networkidle0 instead of the default networkidle2.
Type: object
Default: {headless: true}
Options passed to puppeteer.launch().
Note: Some of the launch options are overridden by the debug option.
To capture screenshots of websites with self-signed certificates, set acceptInsecureCerts: true:
await captureWebsite.file('https://localhost:8080', 'screenshot.png', {
launchOptions: {
acceptInsecureCerts: true
}
});Type: boolean
Default: false
Overwrite the destination file if it exists instead of throwing an error.
This option applies only to captureWebsite.file().
Type: string | Function
Default: undefined
Inject a function to be executed prior to navigation.
This can be useful for altering the JavaScript environment. For example, you could define a global method on the window, overwrite navigator.languages to change the language presented by the browser, or mock Math.random to return a fixed value.
Type: Array
Default: []
Arguments to pass to the preloadFunction.
The arguments must be serializable (JSON-serializable values, not functions or DOM elements).
await captureWebsite.file('https://example.com', 'screenshot.png', {
preloadFunction: (customValue, anotherValue) => {
window.myValue = customValue;
window.myOtherValue = anotherValue;
},
preloadFunctionArguments: ['Hello', 42]
});Type: object
Define the screenshot's position and size (clipping region).
The position can be specified through x and y coordinates which starts from the top-left.
This can be useful when you only need a part of the page.
You can also consider using element option when you have a CSS selector.
Note that clip is mutually exclusive with the element and fullPage options.
- x - X-coordinate where the screenshot starts.
Type:
number - y - Y-coordinate where the screenshot starts.
Type:
number - width - The width of the screenshot.
Type:
number - height - The height of the screenshot.
Type:
number
For example, define the screenshot's width and height to 400 at position (0, 0):
import captureWebsite from 'capture-website';
await captureWebsite.file('https://sindresorhus.com', 'screenshot.png', {
clip: {
x: 0,
y: 0,
width: 400,
height: 400
}
});Type: string[]
Devices supported by the emulateDevice option.
import captureWebsite, {devices} from 'capture-website';
console.log(devices);
//=> ['iPhone 13', 'iPad Pro', …]Avoid unlimited concurrency as each capture creates a new browser instance. Use a concurrency limiter like p-limit:
import pLimit from 'p-limit';
import captureWebsite from 'capture-website';
const limit = pLimit(2); // Limit to 2 concurrent captures
const items = [
['https://sindresorhus.com', 'sindresorhus'],
['https://github.com', 'github'],
// …
];
await Promise.all(items.map(([url, filename]) => {
return limit(() => captureWebsite.file(url, `${filename}.png`));
}));Or use p-map:
import pMap from 'p-map';
import captureWebsite from 'capture-website';
const items = [
['https://sindresorhus.com', 'sindresorhus'],
['https://github.com', 'github'],
// …
];
await pMap(items, async ([url, filename]) => {
await captureWebsite.file(url, `${filename}.png`);
}, {concurrency: 2});Check out filenamify-url if you need to create a filename from the URL.
If you get an error like No usable sandbox! or Running as root without --no-sandbox is not supported, you need to properly set up sandboxing on your Linux instance.
Alternatively, if you completely trust the content, you can disable sandboxing (strongly discouraged):
import captureWebsite from 'capture-website';
await captureWebsite.file('…', '…', {
launchOptions: {
args: [
'--no-sandbox',
'--disable-setuid-sandbox'
]
}
});How is this different from your Pageres project?
The biggest difference is that Pageres supports capturing multiple screenshots in a single call and it automatically generates the filenames and writes the files. Also, when projects are popular and mature, like Pageres, it becomes harder to make drastic changes. There are many things I would change in Pageres today, but I don't want to risk making lots of breaking changes for such a large userbase before I know whether it will work out or not. So this package is a rethink of how I would have made Pageres had I started it today. I plan to bring some things back to Pageres over time.
- capture-website-cli - CLI for this module
- pageres - A different take on screenshotting websites
