browserless is an efficient driver for controlling headless browsers built on top of puppeteer developed for scenarios where performance matters.

• Puppeteer-like API for common tasks (text, screenshot, html, pdf).
• Built-in evasion techniques to prevent being blocked.
• Built-in adblocker for canceling unnecessary requests.
• Shell interaction via Browserless CLI.
• Easy Google Lighthouse integration.
• Automatic retry & error handling.
• Sensible good defaults.

You can install it via npm:

$ npm install browserless puppeteer --save

browserless is backed by puppeteer, so you need to install it as well.

You can use it next to puppeteer, puppeteer-core or puppeteer-firefox, interchangeably.

This is a full example for showcase all the browserless capabilities:

const createBrowserless = require('browserless')
const termImg = require('term-img')

// First, create a browserless factory
// that it will keep a singleton process running
const browserlessFactory = createBrowserless()

// After that, you can create as many browser context
// as you need. The browser contexts won't share cookies/cache
// with other browser contexts.
const browserless = await browserlessFactory.createContext()

// Perform the action you want, e.g., getting the HTML markup
const buffer = await browserless.screenshot('http://example.com', {
  device: 'iPhone 6'
})

console.log(termImg(buffer))

// After your task is done, destroy your browser context
await browserless.destroyContext()

// At the end, gracefully shutdown the browser process
await browserlessFactory.close()

As you can see, browserless is implemented using a single browser process and creating/destroying specific browser contexts.

You can read more about that at technical details section.

If you're already using puppeteer, you can upgrade to use browserless instead almost with no effort.

Additionally, you can use some specific packages in your codebase, interacting with them from puppeteer.

All methods follow the same interface:

• <url>: The target URL. It's required.
• [options]: Specific settings for the method. It's optional.

The methods follows an async interface, returning a Promise.

It initializes a singleton browserless process, returning a factory that will be used for creating browser contexts:

const browserlessFactory = require('browserless')

const { createContext } = browserlessFactory({
  timeout: 25000,
  lossyDeviceName: true,
  ignoreHTTPSErrors: true
})

// Now every time you call `createContext`
// it will be create a browser context.
const browserless = await createContext({ retry: 2 })

They are some propetary browserless options; The rest of options will be passed to puppeter.launch.

See puppeteer.launch#options.

Additionally, you can setup:

type: string
default: 'Macbook Pro 13'

Sets a consistent device viewport for each page.

type: boolean
default: false

It enables lossy detection over the device descriptor input.

const browserless = require('browserless')({ lossyDeviceName: true })

browserless.getDevice({ device: 'macbook pro 13' })
browserless.getDevice({ device: 'MACBOOK PRO 13' })
browserless.getDevice({ device: 'macbook pro' })
browserless.getDevice({ device: 'macboo pro' })

This setting is oriented for find the device even if the descriptor device name is not exactly the same.

type: string
default: launch
values: 'launch' | 'connect'

It defines if browser should be spawned using puppeteer.launch or puppeteer.connect

type: number
default: 30000

This setting will change the default maximum navigation time.

type: Puppeteer
default: puppeteer|puppeteer-core|puppeteer-firefox

It's automatically detected based on your dependencies being supported puppeteer, puppeteer-core or puppeteer-firefox.

Now you have your browserless factory instantiated, you can create browser contexts on demand:

const browserless = browserlessFactory.createContext({
  retry: 2
})

Every browser context is isolated. They won't share cookies/cache with other browser contexts. They also can contain specific options.

Any browser.createIncognitoBrowserContext#options can be passed.

Additionally, you can setup:

type: number
default: 2

The number of retries that can be performed before considering a navigation as failed.

It returns the Browser instance associated with your browserless factory.

const browser = await browserlessFactory.browser()
console.log('My browser PID is', browser.proces().pid)

It will respawn the singleton browser associated with your browserless factory.

const getPID = promise => (await promise).process().pid

console.log('Process PID:', await getPID(browserlessFactory.browser()))

await browserlessFactory.respawn()

console.log('Process PID:', await getPID(browserlessFactory.browser()))

This method is am implementation detail, normally you don't need to call it.

It will close the singleton browser associated with your browserless factory.

const exitHook = require('exit-hook')

exitHook(browserlessFactory.close())

It should be used to gracefully shutdown your resources.

It serializes the content from the target url into HTML.

const html = await browserless.html('https://example.com')
console.log(html)

See browserless.goto to know all the options and values supported.

It serializes the content from the target url into plain text.

const text = await browserless.text('https://example.com')
console.log(text)

See browserless.goto to know all the options and values supported.

It generates the PDF version of a website behind an url.

const buffer = await browserless.pdf('https://example.com')
console.log(`PDF generated in ${buffer.byteLength()} bytes`)

This method use the following options by default:

{
  margin: '0.35cm',
  printBackground: true,
  scale: 0.65
}

See browserless.goto to know all the options and values supported.

Also, any page.pdf option is supported.

Additionally, you can setup:

type: string | string[]
default: '0.35cm'

It sets paper margins. All possible units are:

• px for pixel.
• in for inches.
• cm for centimeters.
• mm for millimeters.

You can pass an object object specifying each corner side of the paper:

const buffer = await browserless.pdf(url.toString(), {
  margin: {
    top: '0.35cm',
    bottom: '0.35cm',
    left: '0.35cm',
    right: '0.35cm'
  }
})

Or, in case you pass an string, it will be used for all the sides:

const buffer = await browserless.pdf(url.toString(), {
  margin: '0.35cm'
})

It takes a screenshot from the target url.

const buffer = await browserless.screenshot('https://example.com')
console.log(`Screenshot taken in ${buffer.byteLength()} bytes`)

This method use the following options by default:

{
  device: 'macbook pro 13'
}

See browserless.goto to know all the options and values supported.

Also, any page.screenshot option is supported.

Additionally, you can setup:

type: string
default: 'atom-dark'

When this value is present and the response 'Content-Type' header is 'json', it beautifies HTML markup using Prism.

The syntax highlight theme can be customized, being possible to setup:

• A prism-themes identifier (e.g., 'dracula').
• A remote URL (e.g., 'https://unpkg.com/prism-theme-night-owl').

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.

type: object

After the screenshot has been taken, this option allows you to place the screenshot into a fancy overlay

You can configure the overlay specifying:

• browser: It sets the browser image overlay to use, being light and dark supported values.
• background: It sets the background to use, being supported to pass:
  • An hexadecimal/rgb/rgba color code, eg. #c1c1c1.
  • A CSS gradient, eg. linear-gradient(225deg, #FF057C 0%, #8D0B93 50%, #321575 100%)
  • An image url, eg. https://source.unsplash.com/random/1920x1080.

const buffer = await browserless.screenshot(url.toString(), {
  styles: [
    '.crisp-client, #cookies-policy { display: none; }'
  ],
  overlay: {
    browser: 'dark',
    background:
      'linear-gradient(45deg, rgba(255,18,223,1) 0%, rgba(69,59,128,1) 66%, rgba(69,59,128,1) 100%)'
  }
})

It will destroy the current browser context

const browserless = await browserlessFactory.createContext({ retry: 0 })

const content = await browserless.html('https://example.com')

await browserless.destroyContext()

Giving a specific device descriptons, this method will be the devices settings for it.

browserless.getDevice({ device: 'Macbook Pro 15' })
// {
//   userAgent: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/62.0.3202.89 Safari/537.36',
//   viewport: {
//     width: 1440,
//     height: 900,
//     deviceScaleFactor: 2,
//     isMobile: false,
//     hasTouch: false,
//     isLandscape: false
//   }
// }

It extends from puppeteer.devices, adding some missing devices there.

type: string

The device descriptor name. It's used to find the rest presets associated with it.

When lossyDeviceName is enabled, a fuzzy search rather than a strict search will be performed in order to maximize getting a result back.

type: object

An extra of viewport settings that will be merged with the device presets.

browserless.getDevice({
  device: 'iPad',
  viewport: {
    isLandscape: true
  }
})

type: object

An extra of headers that will be merged with the device presets.

browserless.getDevice({
  device: 'iPad',
  headers: {
    'user-agent': 'googlebot'
  }
})

It exposes an interface for creating your own evaluate function, passing you the page and response.

The fn will receive page and response as arguments:

const ping = browserless.evaluate((page, response) => ({
  statusCode: response.status(),
  url: response.url(),
  redirectUrls: response.request().redirectChain()
}))

await ping('https://example.com')
// {
//   "statusCode": 200,
//   "url": "https://example.com/",
//   "redirectUrls": []
// }

You don't need to close the page; It will be closed automatically.

Internally, the method performs a browserless.goto, being possible to pass extra arguments as second parameter:

const serialize = browserless.evaluate(page => page.evaluate(() => document.body.innerText), {
  waitUntil: 'domcontentloaded'
})

await serialize('https://example.com')
// '<!DOCTYPE html><html><div>…'

It performs a page.goto with a lot of extra capabilities:

const browserless = require('browserless')

const page = await browserless.page()
const { response, device } = await browserless.goto(page, { url: 'http://example.com' })

Any option passed here will bypass to page.goto.

Additionally, you can setup:

type: array
default: []

It sets the ability to abort requests based on the resource type.

type: boolean
default: true

It enabled the builtin adblocker by Cliqz that aborts unnecessary third party requests associated with ads services.

type: boolean
default: false

Disable CSS animations and transitions, also it sets prefers-reduced-motion consequently.

type: string | string[]

Click the DOM element matching the given CSS selector.

type: string
default: 'macbook pro 13'

It specifies the device descriptor to use in order to retrieve userAgent and viewport.

type: string[]
default: require('@browserless/goto').evasions

It makes your Headless undetectable, preventing to being blocked.

These techniques are used by antibot systems to check if you are a real browser and block any kind of automated access. All the evasion techniques implemented are:

The evasion techniques are enabled by default. You can omit techniques just filtering them:

const createBrowserless = require('browserless')

const evasions = require('@browserless/goto').evasions.filter(
  evasion => evasion !== 'randomizeUserAgent'
)

const browserlessFactory = createBrowserless({ evasions })

type: object

An object containing additional HTTP headers to be sent with every request.

const browserless = require('browserless')

const page = await browserless.page()
await browserless.goto(page, {
  url: 'http://example.com',
  headers: {
    'user-agent': 'googlebot',
    cookie: 'foo=bar; hello=world'
  }
})

This sets visibility: hidden on the matched elements.

type: string

In case you provide HTML markup, a page.setContent avoiding fetch the content from the target URL.

type: boolean
default: true

When it's false, it disables JavaScript on the current page.

type: string
default: 'screen'

Changes the CSS media type of the page using page.emulateMediaType.

type: string | string[]

Injects <script type="module"> into the browser page.

It can accept:

• Absolute URLs (e.g., 'https://cdn.jsdelivr.net/npm/@microlink/[email protected]/src/browser.js').
• Local file (e.g., `'local-file.js').
• Inline code (e.g., "document.body.style.backgroundColor = 'red'").

const buffer = await browserless.screenshot(url.toString(), {
  modules: [
    'https://cdn.jsdelivr.net/npm/@microlink/[email protected]/src/browser.js',
    'local-file.js',
    "document.body.style.backgroundColor = 'red'"
  ]
})

type: string
default: 'no-preference'

Sets prefers-color-scheme CSS media feature, used to detect if the user has requested the system use a 'light' or 'dark' color theme.

type: string | string[]

Injects <script> into the browser page.

It can accept:

• Absolute URLs (e.g., 'https://cdn.jsdelivr.net/npm/@microlink/[email protected]/src/browser.js').
• Local file (e.g., `'local-file.js').
• Inline code (e.g., "document.body.style.backgroundColor = 'red'").

const buffer = await browserless.screenshot(url.toString(), {
  scripts: [
    'https://cdn.jsdelivr.net/npm/[email protected]/dist/jquery.min.js',
    'local-file.js',
    "document.body.style.backgroundColor = 'red'"
  ]
})

Prefer to use modules whenever possible.

type: string

Scroll to the DOM element matching the given CSS selector.

type: string | string[]

Injects <style> into the browser page.

It can accept:

• Absolute URLs (e.g., 'https://cdn.jsdelivr.net/npm/[email protected]/dist/dark.css').
• Local file (e.g., `'local-file.css').
• Inline code (e.g., "body { background: red; }").

const buffer = await browserless.screenshot(url.toString(), {
  styles: [
    'https://cdn.jsdelivr.net/npm/[email protected]/dist/dark.css',
    'local-file.css',
    'body { background: red; }'
  ]
})

type: string

It changes the timezone of the page.

type: string

The target URL.

It will setup a custom viewport, using page.setViewport method.

type:string

Wait a quantity of time, selector or function using page.waitForSelector.

type:number

Wait a quantity of time, selector or function using page.waitForTimeout.

type: string | string[]
default: 'auto'
values: 'auto' | 'load' | 'domcontentloaded' | 'networkidle0' | 'networkidle2'

When to consider navigation succeeded.

If you provide an array of event strings, navigation is considered to be successful after all events have been fired.

Events can be either:

• 'auto': A combination of 'load' and 'networkidle2' in a smart way to wait the minimum time necessary.
• 'load': Consider navigation to be finished when the load event is fired.
• 'domcontentloaded': Consider navigation to be finished when the DOMContentLoaded event is fired.
• 'networkidle0': Consider navigation to be finished when there are no more than 0 network connections for at least 500 ms.
• 'networkidle2': Consider navigation to be finished when there are no more than 2 network connections for at least 500 ms.

It returns the BrowserContext associated with your instance.

const browserContext = await browserless.context()

console.log({ isIncognito: browserContext.isIncognito() })
// => { isIncognito: true }

It returns a standalone Page associated with the current browser context.

const page = await browserless.page()
await page.content()
// => '<html><head></head><body></body></html>'

You can perform any browserless action from your terminal.

Just you need to install @browserless/cli globally:

npm install @browserless/cli --global

Additionally, can do it under demand using npx:

npx @browserless/cli --help

That's the preferred way to interact with the CLI under CI/CD scenarios.

browserless has a Lighthouse integration that connects to a Puppeteer instance in a simple way.

const lighthouse = require('@browserless/lighthouse')
const { writeFile } = require('fs/promises')

const report = await lighthouse('https://example.com')

await writeFile('report.json', JSON.stringify(report, null, 2))

The report will be generated url, extending from lighthouse:default settings, being these settings the same than Google Chrome Audits reports on Developer Tools.

The second argument can contain lighthouse specific settings The following options are used by default:

{
  logLevel: 'error',
  output: 'json',
  device: 'desktop',
  onlyCategories: ['perfomance', 'best-practices', 'accessibility', 'seo']
}

See Lighthouse configuration to know all the options and values supported.

Additionally, you can setup:

type: function
default: require('browserless')

The browserless instance to use for getting the browser.

type: string
default: 'error'
values: 'silent' | 'error' | 'info' | 'verbose'

The level of logging to enable.

type: string | string[]
default: 'json'
values: 'json' | 'csv' | 'html'

The type(s) of report output to be produced.

type: string
default: 'desktop'
values: 'desktop' | 'mobile' | 'none'

How emulation (useragent, device screen metrics, touch) should be applied. 'none' indicates Lighthouse should leave the host browser as-is.

type: string[] | null
default: ['performance', 'best-practices', 'accessibility', 'seo']
values: 'performance' | 'best-practices' | 'accessibility' | 'pwa' | 'seo'

Includes only the specified categories in the final report.

browserless is internally divided into multiple packages for ensuring just use the minimum quantity of code necessary for your use case.

Package	Version
browserless
@browserless/benchmark
@browserless/cli
@browserless/devices
@browserless/examples
@browserless/errors
@browserless/function
@browserless/goto
@browserless/pdf
@browserless/screenshot
@browserless/lighthouse

Q: Why use browserless over puppeteer?

browserless not replace puppeteer, it complements. It's just a syntactic sugar layer over official Headless Chrome oriented for production scenarios.

Q: Why do you block ads scripts by default?

Headless navigation is expensive compared with just fetch the content from a website.

In order to speed up the process, we block ads scripts by default because they are so bloat.

Q: My output is different from the expected

Probably browserless was too smart and it blocked a request that you need.

You can active debug mode using DEBUG=browserless environment variable in order to see what is happening behind the code:

Consider open an issue with the debug trace.

Q: I want to use browserless with my AWS Lambda like project

Yes, check chrome-aws-lambda to setup AWS Lambda with a binary compatible.

browserless © Microlink, released under the MIT License.
Authored and maintained by Microlink with help from contributors.

The logo has been designed by xinh studio.

microlink.io · GitHub microlinkhq · Twitter @microlinkhq