As web developers, we usually enjoy the strong security net of the browser - the risks associated with the code we write are relatively small. Our websites are granted limited powers in a sandbox, and we trust that our users enjoy a browser built by a large team of engineers that is able to quickly respond to newly discovered security threats.
When working with Electron, it is important to understand that Electron is not a web browser. It allows you to build feature-rich desktop applications with familiar web technologies, but your code wields much greater power. JavaScript can access the filesystem, user shell, and more. This allows you to build high quality native applications, but the inherent security risks scale with the additional powers granted to your code.
With that in mind, be aware that displaying arbitrary content from untrusted sources poses a severe security risk that Electron is not intended to handle. In fact, the most popular Electron apps (Atom, Slack, Visual Studio Code, etc) display primarily local content (or trusted, secure remote content without Node integration) – if your application executes code from an online source, it is your responsibility to ensure that the code is not malicious.
For information on how to properly disclose an Electron vulnerability, see SECURITY.md
While Electron strives to support new versions of Chromium as soon as possible, developers should be aware that upgrading is a serious undertaking - involving hand-editing dozens or even hundreds of files. Given the resources and contributions available today, Electron will often not be on the very latest version of Chromium, lagging behind by either days or weeks.
We feel that our current system of updating the Chromium component strikes an appropriate balance between the resources we have available and the needs of the majority of applications built on top of the framework. We definitely are interested in hearing more about specific use cases from the people that build things on top of Electron. Pull requests and contributions supporting this effort are always very welcome.
A security issue exists whenever you receive code from a remote destination and
execute it locally. As an example, consider a remote website being displayed
inside a BrowserWindow. If an attacker somehow manages to
change said content (either by attacking the source directly, or by sitting
between your app and the actual destination), they will be able to execute
native code on the user's machine.
⚠️ Under no circumstances should you load and execute remote code with Node.js integration enabled. Instead, use only local files (packaged together with your application) to execute Node.js code. To display remote content, use the<webview>tag and make sure to disable thenodeIntegration.
From Electron 2.0 on, developers will see warnings and recommendations printed to the developer console. They only show up when the binary's name is Electron, indicating that a developer is currently looking at the console.
You can force-enable or force-disable these warnings by setting
ELECTRON_ENABLE_SECURITY_WARNINGS or ELECTRON_DISABLE_SECURITY_WARNINGS on
either process.env or the window object.
This is not bulletproof, but at the least, you should follow these steps to improve the security of your application.
- Only load secure content
- Disable the Node.js integration in all renderers that display remote content
- Enable context isolation in all renderers that display remote content
- Use
ses.setPermissionRequestHandler()in all sessions that load remote content - Do not disable
webSecurity - Define a
Content-Security-Policyand use restrictive rules (i.e.script-src 'self') - Override and disable
eval, which allows strings to be executed as code. - Do not set
allowRunningInsecureContenttotrue - Do not enable experimental features
- Do not use
enableBlinkFeatures <webview>: Do not useallowpopups<webview>: Verify options and params- Disable or limit navigation
- Disable or limit creation of new windows
Any resources not included with your application should be loaded using a
secure protocol like HTTPS. In other words, do not use insecure protocols
like HTTP. Similarly, we recommend the use of WSS over WS, FTPS over
FTP, and so on.
HTTPS has three main benefits:
- It authenticates the remote server, ensuring your app connects to the correct host instead of an impersonator.
- It ensures data integrity, asserting that the data was not modified while in transit between your application and the host.
- It encrypts the traffic between your user and the destination host, making it more difficult to eavesdrop on the information sent between your app and the host.
// Bad
browserWindow.loadURL('http://my-website.com')
// Good
browserWindow.loadURL('https://my-website.com')<!-- Bad -->
<script crossorigin src="http://cdn.com/react.js"></script>
<link rel="stylesheet" href="http://cdn.com/style.css">
<!-- Good -->
<script crossorigin src="https://cdn.com/react.js"></script>
<link rel="stylesheet" href="https://cdn.com/style.css">It is paramount that you disable Node.js integration in any renderer
(BrowserWindow, BrowserView, or
<webview>) that loads remote content. The goal is to limit the
powers you grant to remote content, thus making it dramatically more difficult
for an attacker to harm your users should they gain the ability to execute
JavaScript on your website.
After this, you can grant additional permissions for specific hosts. For example, if you are opening a BrowserWindow pointed at `https://my-website.com/", you can give that website exactly the abilities it needs, but no more.
A cross-site-scripting (XSS) attack is more dangerous if an attacker can jump out of the renderer process and execute code on the user's computer. Cross-site-scripting attacks are fairly common - and while an issue, their power is usually limited to messing with the website that they are executed on. Disabling Node.js integration helps prevent an XSS from being escalated into a so-called "Remote Code Execution" (RCE) attack.
// Bad
const mainWindow = new BrowserWindow()
mainWindow.loadURL('https://my-website.com')// Good
const mainWindow = new BrowserWindow({
webPreferences: {
nodeIntegration: false,
preload: './preload.js'
}
})
mainWindow.loadURL('https://my-website.com')<!-- Bad -->
<webview nodeIntegration src="page.html"></webview>
<!-- Good -->
<webview src="page.html"></webview>When disabling Node.js integration, you can still expose APIs to your website that
do consume Node.js modules or features. Preload scripts continue to have access
to require and other Node.js features, allowing developers to expose a custom
API to remotely loaded content.
In the following example preload script, the later loaded website will have
access to a window.readConfig() method, but no Node.js features.
const { readFileSync } = require('fs')
window.readConfig = function () {
const data = readFileSync('./config.json')
return data
}Context isolation is an Electron feature that allows developers to run code
in preload scripts and in Electron APIs in a dedicated JavaScript context. In
practice, that means that global objects like Array.prototype.push or
JSON.parse cannot be modified by scripts running in the renderer process.
Electron uses the same technology as Chromium's Content Scripts to enable this behavior.
Context isolation allows each the scripts on running in the renderer to make changes to its JavaScript environment without worrying about conflicting with the scripts in the Electron API or the preload script.
While still an experimental Electron feature, context isolation adds an additional layer of security. It creates a new JavaScript world for Electron APIs and preload scripts.
At the same time, preload scripts still have access to the document and
window objects. In other words, you're getting a decent return on a likely
very small investment.
// Main process
const mainWindow = new BrowserWindow({
webPreferences: {
contextIsolation: true,
preload: 'preload.js'
}
})// Preload script
// Set a variable in the page before it loads
webFrame.executeJavaScript('window.foo = "foo";')
// The loaded page will not be able to access this, it is only available
// in this context
window.bar = 'bar'
document.addEventListener('DOMContentLoaded', () => {
// Will log out 'undefined' since window.foo is only available in the main
// context
console.log(window.foo)
// Will log out 'bar' since window.bar is available in this context
console.log(window.bar)
})You may have seen permission requests while using Chrome: They pop up whenever the website attempts to use a feature that the user has to manually approve ( like notifications).
The API is based on the Chromium permissions API and implements the same types of permissions.
By default, Electron will automatically approve all permission requests unless the developer has manually configured a custom handler. While a solid default, security-conscious developers might want to assume the very opposite.
const { session } = require('electron')
session
.fromPartition('some-partition')
.setPermissionRequestHandler((webContents, permission, callback) => {
const url = webContents.getURL()
if (permission === 'notifications') {
// Approves the permissions request
callback(true)
}
if (!url.startsWith('https://my-website.com')) {
// Denies the permissions request
return callback(false)
}
})Recommendation is Electron's default
You may have already guessed that disabling the webSecurity property on a
renderer process (BrowserWindow,
BrowserView, or <webview>) disables crucial
security features.
Do not disable webSecurity in production applications.
Disabling webSecurity will disable the same-origin policy and set
allowRunningInsecureContent property to true. In other words, it allows
the execution of insecure code from different domains.
// Bad
const mainWindow = new BrowserWindow({
webPreferences: {
webSecurity: false
}
})// Good
const mainWindow = new BrowserWindow()<!-- Bad -->
<webview disablewebsecurity src="page.html"></webview>
<!-- Good -->
<webview src="page.html"></webview>A Content Security Policy (CSP) is an additional layer of protection against cross-site-scripting attacks and data injection attacks. We recommend that they be enabled by any website you load inside Electron.
CSP allows the server serving content to restrict and control the resources
Electron can load for that given web page. https://your-page.com should
be allowed to load scripts from the origins you defined while scripts from
https://evil.attacker.com should not be allowed to run. Defining a CSP is an
easy way to improve your applications security.
Electron respects the Content-Security-Policy HTTP header
and the respective <meta> tag.
The following CSP will allow Electron to execute scripts from the current
website and from apis.mydomain.com.
// Bad
Content-Security-Policy: '*'
// Good
Content-Security-Policy: script-src 'self' https://apis.mydomain.comeval() is a core JavaScript method that allows the execution of JavaScript
from a string. Disabling it disables your app's ability to evaluate JavaScript
that is not known in advance.
The eval() method has precisely one mission: To evaluate a series of
characters as JavaScript and execute it. It is a required method whenever you
need to evaluate code that is not known ahead of time. While legitimate use
cases exist, like any other code generators, eval() is difficult to harden.
Generally speaking, it is easier to completely disable eval() than to make
it bulletproof. Thus, if you do not need it, it is a good idea to disable it.
// ESLint will warn about any use of eval(), even this one
// eslint-disable-next-line
window.eval = global.eval = function () {
throw new Error(`Sorry, this app does not support window.eval().`)
}Recommendation is Electron's default
By default, Electron will not allow websites loaded over HTTPS to load and
execute scripts, CSS, or plugins from insecure sources (HTTP). Setting the
property allowRunningInsecureContent to true disables that protection.
Loading the initial HTML of a website over HTTPS and attempting to load
subsequent resources via HTTP is also known as "mixed content".
Loading content over HTTPS assures the authenticity and integrity
of the loaded resources while encrypting the traffic itself. See the section on
only displaying secure content for more details.
// Bad
const mainWindow = new BrowserWindow({
webPreferences: {
allowRunningInsecureContent: true
}
})// Good
const mainWindow = new BrowserWindow({})Recommendation is Electron's default
Advanced users of Electron can enable experimental Chromium features using the
experimentalFeatures and experimentalCanvasFeatures properties.
Experimental features are, as the name suggests, experimental and have not been enabled for all Chromium users. Furthermore, their impact on Electron as a whole has likely not been tested.
Legitimate use cases exist, but unless you know what you are doing, you should not enable this property.
// Bad
const mainWindow = new BrowserWindow({
webPreferences: {
experimentalFeatures: true
}
})// Good
const mainWindow = new BrowserWindow({})Recommendation is Electron's default
Blink is the name of the rendering engine behind Chromium. As with
experimentalFeatures, the enableBlinkFeatures property allows developers to
enable features that have been disabled by default.
Generally speaking, there are likely good reasons if a feature was not enabled by default. Legitimate use cases for enabling specific features exist. As a developer, you should know exactly why you need to enable a feature, what the ramifications are, and how it impacts the security of your application. Under no circumstances should you enable features speculatively.
// Bad
const mainWindow = new BrowserWindow({
webPreferences: {
enableBlinkFeatures: ['ExecCommandInJavaScript']
}
})// Good
const mainWindow = new BrowserWindow()Recommendation is Electron's default
If you are using <webview>, you might need the pages and scripts
loaded in your <webview> tag to open new windows. The allowpopups attribute
enables them to create new BrowserWindows using the
window.open() method. <webview> tags are otherwise not allowed to create new
windows.
If you do not need popups, you are better off not allowing the creation of
new BrowserWindows by default. This follows the principle
of minimally required access: Don't let a website create new popups unless
you know it needs that feature.
<!-- Bad -->
<webview allowpopups src="page.html"></webview>
<!-- Good -->
<webview src="page.html"></webview>A WebView created in a renderer process that does not have Node.js integration
enabled will not be able to enable integration itself. However, a WebView will
always create an independent renderer process with its own webPreferences.
It is a good idea to control the creation of new <webview> tags
from the main process and to verify that their webPreferences do not disable
security features.
Since <webview> live in the DOM, they can be created by a script running on your
website even if Node.js integration is otherwise disabled.
Electron enables developers to disable various security features that control
a renderer process. In most cases, developers do not need to disable any of
those features - and you should therefore not allow different configurations
for newly created <webview> tags.
Before a <webview> tag is attached, Electron will fire the
will-attach-webview event on the hosting webContents. Use the event to
prevent the creation of webViews with possibly insecure options.
app.on('web-contents-created', (event, contents) => {
contents.on('will-attach-webview', (event, webPreferences, params) => {
// Strip away preload scripts if unused or verify their location is legitimate
delete webPreferences.preload
delete webPreferences.preloadURL
// Disable Node.js integration
webPreferences.nodeIntegration = false
// Verify URL being loaded
if (!params.src.startsWith('https://yourapp.com/')) {
event.preventDefault()
}
})
})Again, this list merely minimizes the risk, it does not remove it. If your goal is to display a website, a browser will be a more secure option.
If your app has no need to navigate or only needs to navigate to known pages, it is a good idea to limit navigation outright to that known scope, disallowing any other kinds of navigation.
Navigation is a common attack vector. If an attacker can convince your app to
navigate away from its current page, they can possibly force your app to open
web sites on the Internet. Even if your webContents are configured to be more
secure (like having nodeIntegration disabled or contextIsolation enabled),
getting your app to open a random web site will make the work of exploiting your
app a lot easier.
A common attack pattern is that the attacker convinces your app's users to interact with the app in such a way that it navigates to one of the attacker's pages. This is usually done via links, plugins, or other user-generated content.
If your app has no need for navigation, you can call event.preventDefault()
in a will-navigate handler. If you know which pages your app
might navigate to, check the URL in the event handler and only let navigation
occur if it matches the URLs you're expecting.
We recommend that you use Node's parser for URLs. Simple string comparisons can
sometimes be fooled - a startsWith('https://google.com') test would let
https://google.com.attacker.com through.
const URL = require('url')
app.on('web-contents-created', (event, contents) => {
contents.on('will-navigate', (event, navigationUrl) => {
const parsedUrl = new URL(navigationUrl)
if (parsedUrl.hostname !== 'my-own-server.com') {
event.preventDefault()
}
})
})If you have a known set of windows, it's a good idea to limit the creation of additional windows in your app.
Much like navigation, the creation of new webContents is a common attack
vector. Attackers attempt to convince your app to create new windows, frames,
or other renderer processes with more privileges than they had before; or
with pages opened that they couldn't open before.
If you have no need to create windows in addition to the ones you know you'll
need to create, disabling the creation buys you a little bit of extra
security at no cost. This is commonly the case for apps that open one
BrowserWindow and do not need to open an arbitrary number of additional
windows at runtime.
webContents will emit the new-window event
before creating new windows. That event will be passed, amongst other
parameters, the url the window was requested to open and the options used to
create it. We recommend that you use the event to scrutinize the creation of
windows, limiting it to only what you need.
const { shell } = require('electron')
app.on('web-contents-created', (event, contents) => {
contents.on('new-window', (event, navigationUrl) => {
// In this example, we'll ask the operating system
// to open this event's url in the default browser.
event.preventDefault()
shell.openExternal(navigationUrl)
})
})