Playwright-with-fingerprints NPM

playwright-with-fingerprints

This is the repo for playwright-with-fingerprints, a plugin for the playwright framework that allows you to change a browser fingerprint, generate a virtual identity and improve your browser's stealth.

In order to achieve this, the FingerprintSwitcher service is used, which allows you to replace a list of important browser properties, and thus you will act like a completely new user.

Warning: plugin is still in beta stage, it means that bugs may happen, including critical.

Adding a plugin to your project is very easy - it only takes a few lines of code. You just need to change the browser startup code a bit and add method calls to get and apply fingerprints. The rest of the code can remain unchanged. In general, only four basic steps are required, see the example below:

https://user-images.githubusercontent.com/30115373/198844008-de84d5f8-db14-498d-a7f0-630da931177e.mp4

Current supported engine version - 123.0.6312.59.

About

This library allows you to change browser fingerprint and use playwright automation framework with enhanced anonymity. In order to migrate and use it, a minimum of modifications and code changes is required.

Browser fingerprinting - is a technique that allows to identify the user by a combination of browser properties, such as - fonts, resolution, list of plugins, navigator properties, etc. By adding new factors and using browser API in a special way, a site can determine exactly which user is visiting it, even if the user is using a proxy. When using this package and replacing fingerprints, websites will not be able to identify you from other users, as all these properties and results of API calls will be replaced with values from real devices.

Let's look at a small example of WebGL property substitution. In the screenshot below, the left column shows the values from the regular browser, and the right column shows the values substituted using ready-made fingerprints. This result cannot be achieved using only the replacement of various browser properties via JavaScript, that's what this plugin and service is for:

WebGL

You can learn more by following this link.

Installation

To use this plugin in your project, install it with your favorite package manager:

npm i playwright-with-fingerprints
# or
pnpm i playwright-with-fingerprints
# or
yarn add playwright-with-fingerprints

The playwright-core or playwright packages must also be installed. If one of these packages is already installed, you can skip this step:

npm i playwright
# or
npm i playwright-core

Both options will work correctly and do not require additional steps to make the plugin work with them. But keep in mind that some versions may not be supported. In this case, you will get an error when importing the plugin. You can always find supported version numbers here.

Please note that this plugin only supports the default playwright library, wrappers have not been tested and may cause errors. Also, according to the architecture section, this plugin can only be installed and used on Windows operating system.

Creating new project

Here's how to start working on a project from scratch.

First you need to import the playwright-with-fingerprints library:

const { plugin } = require('playwright-with-fingerprints');

No need to require the playwright or playwright-core.

After that, you need to obtain the fingerprint from the server and apply it:

const fingerprint = await plugin.fetch('', {
  tags: ['Microsoft Windows', 'Chrome'],
});

plugin.useFingerprint(fingerprint);

When this code is executed, the fingerprint variable will contain the data required to apply the fingerprint. It's a string, you can save it to a file and use it later. Running this code for the first time can be very slow. Additional time is needed to download the browser.

Finally, you need to create the browser instance:

const browser = await plugin.launch();

The parameters of the launch method are the same as the corresponding method in the playwright library, link. The browser variable will contain an instance of the Browser class defined in the playwright library. It means that it can be used to write a new or use an existing playwright script without any changes. You need to rely on the documentation of the original framework on how to use control browser - link.

Here is the complete code, you can copy/paste it and try:

const { plugin } = require('playwright-with-fingerprints');

(async () => {
  // Get a fingerprint from the server:
  const fingerprint = await plugin.fetch('', {
    tags: ['Microsoft Windows', 'Chrome'],
  });

  // Apply fingerprint:
  plugin.useFingerprint(fingerprint);

  // Launch the browser instance:
  const browser = await plugin.launch();

  // The rest of the code is the same as for a standard `playwright` library:
  const page = await browser.newPage();
  await page.goto('https://example.com');

  // Print the browser viewport size:
  console.log(
    'Viewport:',
    await page.evaluate(() => ({
      deviceScaleFactor: window.devicePixelRatio,
      width: document.documentElement.clientWidth,
      height: document.documentElement.clientHeight,
    }))
  );

  await browser.close();
})();

Also take a look at the TypeScript declarations here for more details about the exported classes, methods, and properties. Thanks to them, when using the library, auto-completion with detailed descriptions will work.

Migration

There are a few steps required to start using fingerprints for your existing project:

Import playwright-with-fingerprints instead of playwright or playwright-core.
Call the useFingerprint and/or useProxy methods to apply the fingerprint and proxy before starting the browser.
Launch the browser using the plugin.launch method (the plugin variable was imported in the first step).
The rest of the code can be left unchanged.

Consider you have the following project using the playwright:

/* Without fingerprints */
const { chromium } = require('playwright');

(async () => {
  const browser = await chromium.launch();

  const page = await browser.newPage();
  await page.goto('https://browserleaks.com/canvas', { waitUntil: 'networkidle' });

  console.log('Canvas signature:', await page.$eval('#crc', (el) => el.innerText));

  await browser.close();
})();

It verifies the canvas signature by visiting this URL and parsing the corresponding value. No matter how many times you run this test on the same machine, the results will be the same. This is because this test depends on the hardware you are using - if the hardware stays the same, the results will also be the same. But if we change the fingerprint, the results will be different.

Let's modify this project to add fingerprint support. The updated code will look like this:

/* With fingerprints */

// Import `playwright-with-fingerprints` instead of `playwright`
// const { chromium } = require('playwright');
const { plugin } = require('playwright-with-fingerprints');

(async () => {
  // Obtain a fingerprint from the server. The resulting variable contains a string - it can be stored for later use:
  const fingerprint = await plugin.fetch('', {
    tags: ['Microsoft Windows', 'Chrome'],
  });

  // Apply fingerprint - after calling the `useFingerprint` method, the browser will be launched with a fingerprint:
  plugin.useFingerprint(fingerprint);

  // Replace `chromium.launch` method call with `plugin.launch`:
  // const browser = await chromium.launch();
  const browser = await plugin.launch();

  // The rest of the code is the same as for the standard `playwright` library:
  const page = await browser.newPage();
  await page.goto('https://browserleaks.com/canvas', { waitUntil: 'networkidle' });

  console.log('Canvas signature:', await page.$eval('#crc', (el) => el.innerText));

  await browser.close();
})();

After running the updated code, a new fingerprint will be applied each time, so the scores will be different for each run.

Launching the browser

You can launch the browser in two different ways. There are two methods for this - launch and spawn.

The parameters and return type of the launch method are exactly the same as for BrowserType.launch method. You can use the official API documentation to learn more about them. The launch method also has the same purpose - to start a new browser instance with the given parameters and connect to it.

In addition to the standard functionality, it allows you to change the fingerprint and proxy using the useFingerprint and useProxy methods. A detailed description and annotations can also be found here.

const { plugin } = require('playwright-with-fingerprints');

const browser = await plugin.launch({
  args: ['--mute-audio'],
  headless: true,
});

The spawn method works in a similar way, but uses a separate mechanism to launch the browser. The main difference is that this method just starts the process, but doesn't connect to it for automation - you can do it yourself later.

This method returns a running browser instance that can be used to connect to an existing session using playwright:

const { chromium } = require('playwright');
const { plugin } = require('playwright-with-fingerprints');

const chrome = await plugin.spawn({ headless: true });

const browser = await chromium.connect(chrome.url);

await browser.close();
await chrome.close();

At this link you can find a detailed description of all the options allowed for spawn method. The same goes for the return type declaration, details of which can be found here.

If possible, use it only in extreme cases. It is much more convenient to use the launch method to launch the browser, which minimizes the number of steps for proper initialization and configuration.

Annotations are described for all plugins methods directly in the library code via the TypeScript declarations, so when using it you will be able to see hints for all options and types. You can also find out about it directly here and here.

Configuring plugin

At the moment, it is possible to change the working folder and timeout for requests to the engine, which is used when fetching, applying fingerprints, and so on:

// Set the folder where the plugin engine will be installed:
plugin.setWorkingFolder('./engine');

// Set the timeout used when fetching fingerprints and so on:
plugin.setRequestTimeout(5 * 60000);

The methods from the example above change the settings globally, that is, for all instances of the plugin.

The default values are the ./data directory for the working folder and 300000 milliseconds for the request timeout.

The same result can be achieved using environment variables, however it is strongly recommended to use the described methods.

Configuring browser

In order to change the fingerprint and proxy for your browser, you should use special separate methods:

useProxy - to change the proxy configuration.
useProfile - to change the profile configuration.
useFingerprint - to change the fingerprint configuration.

These methods directly affect only the next launch of the browser. So you should always use them before using the launch plugin method.

You cannot change the settings once the browser is launched - more specifically, an already launched instance will not be affected by the new configuration. But you can safely change the options for the next run, or for a separate browser instance with a different unique configuration.

You can also chain calls, since all these methods return the current plugin instance. It does not matter in which order the settings will be applied. It might look like this:

const { plugin } = require('playwright-with-fingerprints');

const fingerprint = await plugin.fetch('', {
  tags: ['Microsoft Windows', 'Chrome'],
});

plugin.useProxy('127.0.0.1:8080').useFingerprint(fingerprint);

Use these links to see a detailed description of the methods:

This one for the useFingerprint method (also see additional options here).
This one for the useProfile method (also see additional options here).
This one for the useProxy method (also see additional options here).

The usage of these methods is very similar - each takes two parameters, the first of which is the configuration data itself, and the second is additional options. The fingerprint and proxy will not be changed unless the appropriate method is used. In this case, all settings related to browser fingerprinting will remain at their original values.

Fingerprint and proxy aren't applied instantly when calling methods. Instead, the configuration is saved and used directly when the browser is launched using the launch or spawn methods. Thus, you can pre-configure the plugin in a certain way, or change something immediately before launching the browser.

Configuring browser version

You can change the browser version right while using the plugin - the engine may come with several different builds of the browser.

In order to do this, use the useBrowserVersion method or the version property (deprecated). The last one defaults to default, which means that the latest available version will be used:

const { plugin } = require('playwright-with-fingerprints');

// Use a specific version:
plugin.useBrowserVersion('115.0.5790.99');

// Use the latest available version:
plugin.useBrowserVersion('default');

If you specify an unavailable or invalid version, an appropriate error will be thrown when the browser starts. Also keep in mind that this property only affects a specific instance of the plugin, not the entire library.

In order to get a list of available versions shipped with the engine, use the versions method. It returns a list of browser versions as strings or objects with additional information, depending on the format passed:

const { plugin } = require('playwright-with-fingerprints');

// The list of versions is always sorted in descending order:
await plugin.versions('extended').then((versions) => {
  // The latest available browser version will be used:
  plugin.useBrowserVersion(versions[0]['browser_version']);
});

Thanks to this, you can, for example, use the version that corresponds to a certain fingerprint, and vice versa.

Fingerprint usage

In order to change the fingerprint, you need to run the useFingerprint method before starting the browser, i.e. before using the plugin's launch method.

The useFingerprint method takes two parameters. The first is a string with fingerprint data that you can request from the service. The second is additional options for applying a fingerprint, most of which are applied automatically - for example, the safe replacement of the BatteryAPI and AudioAPI properties:

const { plugin } = require('playwright-with-fingerprints');

const fingerprint = await plugin.fetch('', {
  tags: ['Microsoft Windows', 'Chrome'],
});

plugin.useFingerprint(fingerprint, {
  // It's disabled by default.
  safeElementSize: true,
  // It's enabled by default.
  safeBattery: false,
});

In order to obtain fingerprints you should use the fetch plugin method. Pass the service key as the first argument and additional parameters as the second, if necessary:

const { plugin } = require('playwright-with-fingerprints');

const fingerprint = await plugin.fetch('SERVICE_KEY', {
  tags: ['Microsoft Windows', 'Chrome'],
  // Fetch fingerprints only with a browser version higher than 115:
  minBrowserVersion: 115,
  // Fetch fingerprints only with a browser version lower than 119:
  maxBrowserVersion: 119,
  // Fetch fingerprints only collected in the last 15 days:
  timeLimit: '15 days',
});

All possible settings for fetch method, as well as their descriptions, you can find here.

The special current value can be used to filter fingerprints by browser version - in this case, the version installed for the plugin will be used. It can be very convenient as the browser and fingerprint versions will be exactly the same and you don't have to enter the exact values in multiple places.

You can reuse fingerprints instead of requesting new ones each time. To do this, you can save them to a file or to a database - use any option convenient for you. In this way, you can speed up the process of launching the browser with the parameters you need, organize your storage, filter and sort fingerprints locally, and much more:

const { readFile, writeFile } = require('fs/promises');
const { plugin } = require('playwright-with-fingerprints');

// Save the fingerprint to a file:
const fingerprint = await plugin.fetch('', {
  tags: ['Microsoft Windows', 'Chrome'],
});
await writeFile('fingerprint.json', fingerprint);

// Load fingerprint from file at next run:
plugin.useFingerprint(await readFile('fingerprint.json', 'utf8'));

You can learn more about the options directly when adding these methods - just use the built-in annotations.

You can use any tags, filters (e.g. time limit) and settings if you have a service key.

If you specify an empty string as the first argument, the free version will be used. For a free version you won't be able to use other tags than the default ones, as well as some other filters:

const fingerprint = await plugin.fetch('', {
  // You can only use these tags with the free version:
  tags: ['Microsoft Windows', 'Chrome'],
  // You also cannot use such filters in the free version:
  // minBrowserVersion: 115,
});

In the free version, the PerfectCanvas technology is also not available. There are other limitations when using the free version - for example, limiting the number of requests in a certain period of time. To see the differences and limits of different versions, visit this website.

You can buy a key here to avoid limitations.

Profile usage

In order to use a specific browser profile, you can, among other options, use the useProfile method. As the first parameter, it takes the path to the profile folder - the same value that you specify, for example, for the user-data-dir argument. The second parameter is additional options that are primarily responsible for loading fingerprint and proxy data from the profile folder:

const path = require('path');
const { plugin } = require('playwright-with-fingerprints');

plugin.useProfile(path.resolve('./profile'), {
  // Don't load fingerprint from profile folder:
  loadFingerprint: false,
  // Don't load proxy from profile folder:
  loadProxy: false,
});

By default, the plugin will load proxy and fingerprint data from the profile folder, if they are written there. You can change this behavior by setting the options to false - for example, if you are not sure that the stored proxy is working. Please note that if you add your fingerprint or proxy through the appropriate methods, then the data you specified will be used regardless of the options.

You can also use other options to set the path to the profile folder, such as the userDataDir option, if you don't need additional settings:

const path = require('path');
const { plugin } = require('playwright-with-fingerprints');

const browser = await plugin.launch({
  userDataDir: './profile',
  // Browser arguments can be used as well:
  args: [`--user-data-dir=${path.resolve('./profile')}`],
});

After launching a browser with your profile, the fingerprint and proxy data you specified will always be stored in the profile folder. This setting itself is saved between browser launches, that is, it behaves in the same way as other similar methods. To run different profiles, you need to call this method again with different values for the profile directory.

You can learn more about the parameters and additional options for this method here and here.

Temporary profiles

Since version 1.3.0 the plugin uses its own logic to configure browser profiles. This primarily applies to situations where you don't specify options like user-data-dir yourself.

Each framework makes temporary profiles in such cases in completely different places, but for the convenience and correct operation of some fingerprint components, a different solution is used. If you do not specify the path to the profile yourself, then the engine will create its own temporary one - such a profile will be located directly in the engine folder, for example:

const { plugin } = require('playwright-with-fingerprints');

// The profile will be located in the `data/profiles/UNIQUE_ID` folder:
const browser = await plugin.launch();

If you specify the path yourself, then everything will work as usual - just the plugin will add additional components, like widevine or default extensions, to the specified profile.

Also keep in mind that if you specify your own folder for the engine using the setWorkingFolder method, then the profile path will be adjusted.

Proxy usage

In order to set up a proxy, you should use the useProxy method. The first parameter of this method is a string with information about the proxy. The second parameter is additional options that will be applied to the browser, for example, automatic change of language and time zone:

const { plugin } = require('playwright-with-fingerprints');

plugin.useProxy('127.0.0.1:8080', {
  // Change browser timezone according to proxy:
  changeTimezone: true,
  // Replace browser geolocation according to proxy:
  changeGeolocation: true,
});

You can learn more about the parameters and additional options for this method here and here.

The browser supports two types of proxies - https and socks5. It is better to always specify the proxy type in the address line - otherwise, https will be used by default.

You can use aliases - http instead of https and socks instead of socks5. Proxies with authorization (with login and password) are also supported.

In general, when specifying addresses, you can use many different formats, for example:

127.0.0.1:8080
https://127.0.0.1:8080
socks5://127.0.0.1:8181
username:password@127.0.0.1:8080
socks:127.0.0.1:8080:username:password
https://username:password:127.0.0.1:8080

In order to preserve compatibility with the original library syntax, the proxy can be obtained from the arguments you specified. The proxy-server option will be used as the value, and all other options will be set to their default values. But this will be done only if you didn't call the appropriate method for the proxy configuration:

const { plugin } = require('playwright-with-fingerprints');

const browser = await plugin.launch({
  // The syntax for specifying an argument value
  // is exactly the same as for using a separate method.
  args: ['--proxy-server=https://127.0.0.1:8080'],
});

It's better to replace such code with the useProxy method. This is much more convenient because you can immediately set the additional options you need.

More info

If you are having problems with the default plugin, or want to create multiple instances with different settings, you can use a separate exported createPlugin method. It allows you to create a standalone plugin instance that is used in the same way as a standard one. It takes a playwright compatible launcher object as a parameter:

const { chromium } = require('playwright');
const { createPlugin } = require('playwright-with-fingerprints');

const plugin1 = createPlugin({
  // This method is required, must return the same
  // type and take the same parameters as the built-in.
  launch: (options) => chromium.launch(options),
});

// The default instance is created in the same way.
const plugin2 = createPlugin(chromium);

Use this with caution and only in extreme cases - for example, if you are using a wrapper library. It is much safer to work with a default plugin instance.

If you want to learn more about fingerprint substitution technology, explore the list of replaceable properties and various options, such as tags, get or configure your service key, use this link. There you can also get a test fingerprint and see ready-made values that can be applied to your browser.

More examples

Please take a look at the examples directory with ready-made real code examples. You can run them yourself by cloning the repository locally and installing the dependencies.

Architecture

This plugin uses the FingerprintSwitcher service to get fingerprints. The resulting fingerprints are used later directly when working with the browser and are applied in a special way using a custom configuration files.

There are some limitations in using the package, which may be critical or non-critical depending on your task. For example, for the correct operation of the fingerprint substitution technology, a custom browser with various patches is required. This browser is downloaded and installed automatically the first time the API methods are called. It also implies a limitation - it will be impossible to use standard and other various engines.

Fingerprints aren't generated, but downloaded from the service, as they are collected from real devices. This greatly improves the anonymity and quality of the substitution of various properties.

Also keep in mind that this package only work on the Windows operating system. If you install or run it on other platforms, you will get the corresponding errors. This is a forced measure due to the presence of some critical Windows-only dependencies without which this implementation will not work.

The plugin architecture can be summarized as the following diagram:

Architecture

All packages can only work with the Chrome browser, which comes bundled with the libraries and loads automatically. The path to the executable file is defined on the plugin side and cannot be changed. It means that you will not be able to use not only other versions of Chrome or Chromium, but also other browser engines. The same goes for some framework-specific launch options.

This library tries to replicate the interfaces of the playwright framework as much as possible. Thus, it's convenient to use it not only for new projects, but also when migrating from the original version to this plugin. For things not related to launching the browser and the plugin directly, it's better to use the methods and properties of the original library.

Limitations

Please note that there are some restrictions at the moment:

Only Windows operating system is supported.
Parallel launch of browsers is synchronized between calls.
Working with workers is possible only when specifying a separate working folder for each of them.

Also, there is no guarantee that each of these items will be changed in the future.

Alternatives

Check out other ready-made plugins for popular automation frameworks that have a similar API and architecture:

Plugin for selenium - selenium-with-fingerprints
Plugin for puppeteer - puppeteer-with-fingerprints

Also check out BAS - a great alternative to automate the Chrome browser without programming skills. It also supports fingerprint substitution, has simple and powerful multithreading and other advantages.

Documentation

Here you can find a brief description of methods and classes, as well as links to them.

Tag

Describes a tag value that can be used to filter fingerprints.

Time

Describes a time limit that can be used to filter fingerprints.

Version

Describes an object that provides complete information about the available browser version.

plugin.version

Type: string

Get or set the current browser version used by the plugin instance.

Initially it is set to default, which means that the latest available version will be used.

createPlugin(launcher)

launcher Launcher Playwright (or API compatible) browser launcher.

Returns: PlaywrightFingerprintPlugin A new separate plugin instance.

Create a separate plugin instance using the provided playwright compatible browser launcher.

plugin.versions(format?)

format string? The output format of the returned result.

Returns: Promise<Version[]|string[]> The list of objects with detailed version information, or a list of strings.

Get a list of all available browser versions.

plugin.spawn(options?)

options Options? Launcher options that only apply to the browser when using the spawn method.

Returns: Promise<Browser> Promise which resolves to a browser instance.

Launches a browser instance with given arguments and options when specified.

plugin.launch(options?)

options Playwright.LaunchOptions Set of configurable options to set on the browser.

Returns: Promise<Playwright.Browser> Promise which resolves to a browser instance.

Launches playwright and launches a browser instance with given arguments and options when specified.

plugin.fetch(key, options?)

key string Service key for obtaining a fingerprint.
options FetchOptions? Set of configurable options for getting a browser fingerprint.

Returns: Promise<string> Promise which resolves to a fingerprint string.

Obtain a fingerprint using the specified service key and additional options.

plugin.useBrowserVersion(version)

value string Version value as a string.

Returns: this The same plugin instance with an updated settings (for optional chaining).

Set the current browser version used by the plugin instance.

plugin.useProxy(value?, options?)

value string? Proxy value as a string.
options ProxyOptions? Set of configurable options for applying a proxy.

Returns: this The same plugin instance with an updated settings (for optional chaining).

Set the proxy settings using the specified proxy as a string and additional options when specified.

plugin.useFingerprint(value?, options?)

value string? Fingerprint value as a string.
options FingerprintOptions? Set of configurable options for applying a fingerprint.

Returns: this The same plugin instance with an updated settings (for optional chaining).

Set the fingerprint settings using the specified fingerprint as a string and additional options when specified.

plugin.setWorkingFolder(folder)

folder string The selected working folder.

Change the working folder that the plugin uses to work with the engine.

plugin.setRequestTimeout(timeout)

timeout number The selected request timeout.

Change the request timeout that the plugin uses to work with the engine.

Troubleshooting

If you encounter any issue or bug, please use the issues section of the repository.

Please describe the problem in as much detail as possible when creating tickets - indicate the sequence of actions (steps) to repeat the problem, error output, and so on.

If the code is large, attach it as files or use sandboxes. At the same time, it's better to remove from it areas that do not relate to the problem - it will be much easier to figure it out. Format your code and wrap it in special markdown tags if you're adding it to an issue report, for example:

// your code

Please be careful not to attach various secrets in your code or screenshots - for example, fingerprint service keys, account passwords, and so on. In the case of service keys, this can lead to their blocking without a refund.

If the recommendations are not followed, your ticket may be ignored.

Testing

The excellent mocha framework is used for tests in this library. Use the command line or ready-made scripts if you want to run them yourself.

You can also use the FINGERPRINT_CWD environment variable to specify the directory where the engine will be stored, for example:

FINGERPRINT_CWD="../plugin-engine"

The FINGERPRINT_TIMEOUT variable can be set if it's necessary to change the default timeout for executing engine methods, such as applying or fetching a fingerprint: