How to build Mailgun Email passwordless sign-in with Chrome extension

For our new friends:

Logto is an Auth0 alternative designed for modern apps and SaaS products. It offers both Cloud and Open-source services to help you quickly launch your identity and management (IAM) system. Enjoy authentication, authorization, and multi-tenant management all in one.

We recommend starting with a free development tenant on Logto Cloud. This allows you to explore all the features easily.

In this article, we will go through the steps to quickly build the Mailgun sign-in experience (user authentication) with Chrome extension and Logto.

Prerequisites

A running Logto instance. Check out the introduction page to get started.
Basic knowledge of Chrome extension.
A usable Mailgun account.

Create an application in Logto

Logto is based on OpenID Connect (OIDC) authentication and OAuth 2.0 authorization. It supports federated identity management across multiple applications, commonly called Single Sign-On (SSO).

To create your Single page app application, simply follow these steps:

Open the Logto Console. In the "Get started" section, click the "View all" link to open the application frameworks list. Alternatively, you can navigate to Logto Console > Applications, and click the "Create application" button.
In the opening modal, click the "Single page app" section or filter all the available "Single page app" frameworks using the quick filter checkboxes on the left. Click the "Chrome extension" framework card to start creating your application.
Enter the application name, e.g., "Bookstore," and click "Create application".

🎉 Ta-da! You just created your first application in Logto. You'll see a congrats page which includes a detailed integration guide. Follow the guide to see what the experience will be in your application.

Integrate Chrome extension with Logto

tip:

The following demonstration was tested on Chrome v123.0.6312.87 (arm64). Other versions should also work, as long as they support the chrome APIs used in the SDK.
The sample project is available on our GitHub repository.

Installation

npm
Yarn
pnpm

npm i @logto/chrome-extension

yarn add @logto/chrome-extension

pnpm add @logto/chrome-extension

The authentication flow

Assuming you put a "Sign in" button in your Chrome extension's popup, the authentication flow will look like this:

For other interactive pages in your extension, you just need to replace the Extension popup participant with the page's name. In this tutorial, we will focus on the popup page.

This authentication process follows the OpenID Connect (OIDC) protocol, and Logto enforces strict security measures to protect user sign-in.
If you have multiple apps, you can use the same identity provider (Logto). Once the user signs in to one app, Logto will automatically complete the sign-in process when the user accesses another app.

To learn more about the rationale and benefits of redirect-based sign-in, see Logto sign-in experience explained.

Update the `manifest.json`

Logto SDK requires the following permissions in the manifest.json:

manifest.json
{
  "permissions": ["identity", "storage"],
  "host_permissions": ["https://*.logto.app/*"]
}

permissions.identity: Required for the Chrome Identity API, which is used to sign in and sign out.
permissions.storage: Required for storing the user's session.
host_permissions: Required for the Logto SDK to communicate with the Logto APIs.

note:

If you are using a custom domain on Logto Cloud, you need to update the host_permissions to match your domain.

Set up a background script (service worker)

In your Chrome extension's background script, initialize the Logto SDK:

service-worker.js
import LogtoClient from '@logto/chrome-extension';

export const logtoClient = new LogtoClient({
  endpoint: '<your-logto-endpoint>'
  appId: '<your-logto-app-id>',
});

Replace <your-logto-endpoint> and <your-logto-app-id> with the actual values. You can find these values in the application page you just created in the Logto Console.

If you don't have a background script, you can follow the official guide to create one.

info:

Why do we need a background script?

Normal extension pages like the popup or options page can't run in the background, and they have the possibility to be closed during the authentication process. A background script ensures the authentication process can be properly handled.

Then, we need to listen to the message from other extension pages and handle the authentication process:

service-worker.js
chrome.runtime.onMessage.addListener((message, sender, sendResponse) => {
  // In the below code, since we return `true` for each action, we need to call `sendResponse`
  // to notify the sender. You can also handle errors here, or use other ways to notify the sender.

  if (message.action === 'signIn') {
    const redirectUri = chrome.identity.getRedirectURL('/callback');
    logtoClient.signIn(redirectUri).finally(sendResponse);
    return true;
  }

  if (message.action === 'signOut') {
    const redirectUri = chrome.identity.getRedirectURL();
    logtoClient.signOut(redirectUri).finally(sendResponse);
    return true;
  }

  return false;
});

You may notice there are two redirect URIs used in the code above. They are both created by chrome.identity.getRedirectURL, which is a built-in Chrome API to generate a redirect URL for auth flows. The two URIs will be:

https://<extension-id>.chromiumapp.org/callback for sign-in.
https://<extension-id>.chromiumapp.org/ for sign-out.

Note that these URIs are not accessible, and they are only used for Chrome to trigger specific actions for the authentication process.

Update Logto application settings

Now we need to update the Logto application settings to allow the redirect URIs we just created.

Go to the application page in the Logto Console.
In the "Redirect URIs" section, add the URI: https://<extension-id>.chromiumapp.org/callback.
In the "Post sign-out redirect URIs" section, add the URI: https://<extension-id>.chromiumapp.org/.
In the "CORS allowed origins" section, add the URI: chrome-extension://<extension-id>. The SDK in Chrome extension will use this origin to communicate with the Logto APIs.
Click on Save changes.

Remember to replace <extension-id> with your actual extension ID. You can find the extension ID in the chrome://extensions page.

We're almost there! Let's add the sign-in and sign-out buttons and other necessary logic to the popup page.

In the popup.html file:

popup.html
<button id="sign-in">Sign in</button> <button id="sign-out">Sign out</button>

In the popup.js file (assuming popup.js is included in the popup.html):

popup.js
document.getElementById('sign-in').addEventListener('click', async () => {
  await chrome.runtime.sendMessage({ action: 'signIn' });
  // Sign-in completed (or failed), you can update the UI here.
});

document.getElementById('sign-out').addEventListener('click', async () => {
  await chrome.runtime.sendMessage({ action: 'signOut' });
  // Sign-out completed (or failed), you can update the UI here.
});

Checkpoint: Test the authentication flow

Now you can test the authentication flow in your Chrome extension:

Open the extension popup.
Click on the "Sign in" button.
You will be redirected to the Logto sign-in page.
Sign in with your Logto account.
You will be redirected back to the Chrome.

Check authentication state

Since Chrome provide unified storage APIs, rather than the sign-in and sign-out flow, all other Logto SDK methods can be used in the popup page directly.

In your popup.js, you can reuse the LogtoClient instance created in the background script, or create a new one with the same configuration:

popup.js
import LogtoClient from '@logto/chrome-extension';

const logtoClient = new LogtoClient({
  endpoint: '<your-logto-endpoint>'
  appId: '<your-logto-app-id>',
});

// Or reuse the logtoClient instance created in the background script
import { logtoClient } from './service-worker.js';

Then you can create a function to load the authentication state and user's profile:

popup.js
const loadAuthenticationState = async () => {
  const isAuthenticated = await logtoClient.isAuthenticated();
  // Update the UI based on the authentication state

  if (isAuthenticated) {
    const user = await logtoClient.getIdTokenClaims(); // { sub: '...', email: '...', ... }
    // Update the UI with the user's profile
  }
};

You can also combine the loadAuthenticationState function with the sign-in and sign-out logic:

popup.js
document.getElementById('sign-in').addEventListener('click', async () => {
  await chrome.runtime.sendMessage({ action: 'signIn' });
  await loadAuthenticationState();
});

document.getElementById('sign-out').addEventListener('click', async () => {
  await chrome.runtime.sendMessage({ action: 'signOut' });
  await loadAuthenticationState();
});

Here's an example of the popup page with the authentication state:

Other considerations

Service worker bundling: If you use a bundler like Webpack or Rollup, you need to explicitly set the target to browser or similar to avoid unnecessary bundling of Node.js modules.
Module resolution: Logto Chrome extension SDK is an ESM-only module.

See our sample project for a complete example with TypeScript, Rollup, and other configurations.

Add Mailgun connector

Email connector is a method used to send one-time passwords (OTPs) for authentication. It enables Email address verification to support passwordless authentication, including Email-based registration, sign-in, two-factor authentication (2FA), and account recovery. You can easily connect Mailgun as your Email provider. With the Logto Email connector, you can set this up in just a few minutes.

To add a Email connector, simply follow these steps:

Navigate to Console > Connector > Email and SMS connectors.
To add a new Email connector, click the "Set up" button and select "Mailgun".
Review the README documentation for your selected provider.
Complete the configuration fields in the "Parameter Configuration" section.
Customize the Email template using the JSON editor.
Test your configuration by sending a verification code to your Email address.

note:

If you are following the in-place Connector guide, you can skip the next section.

Set up Mailgun email connector

Prerequisites

A Mailgun account
An API key from your Mailgun account, requires the permission to send messages (emails). See Where Can I Find My API Key and SMTP Credentials? for more information.

Basic configuration

Fill out the endpoint field when you are using a different Mailgun API endpoint, for example, EU region should be https://api.eu.mailgun.net. The default value is https://api.mailgun.net.
Fill out the domain field with the domain you have registered in your Mailgun account. This value can be found in the Domains section of the Mailgun dashboard. The domain should be in the format example.com, without the https:// or http:// prefix.
Fill out the apiKey field with the API key you have generated in your Mailgun account.
Fill out the from field with the email address you want to send emails from. This email address must be registered in your Mailgun account. The email address should be in the format Sender Name \<[email protected]>.

Deliveries

Config object

The "Deliveries" section allows you to configure the content of the emails to be sent in different scenarios. It is a JSON key-value map where the key is the usage type and the value is an object containing the content config for the email to be sent.

{
  "<usage-type>": {
    // ...
  }
}

Usage types

The following usage types are supported:

Register: The email to be sent when a user is registering.
SignIn: The email to be sent when a user is signing in.
ForgotPassword: The email to be sent when a user is resetting their password.
Generic: The email to be sent when a user is performing a generic action, for example, testing the email connector.

note:

If the usage type is not specified in the deliveries config, the generic email will be sent. If the generic email is not specified, the connector will return an error.

Content config

The connector supports both direct HTML content and Mailgun template. You can use one of them for each usage type.

In both subject and content, you can use the {{code}} placeholder to insert the verification code.

To use direct HTML content, fill out the following fields:

subject: The subject of the email to be sent.
replyTo: The email address to be used as the reply-to address.
html: (Required) The HTML content of the email to be sent.
text: The plain text version of the email to be sent.

To use Mailgun template, fill out the following fields:

subject: The subject of the email to be sent.
replyTo: The email address to be used as the reply-to address.
template: (Required) The name of the Mailgun template to be used.
variables: The variables to be passed to the Mailgun template. Should be a JSON key-value map since it will be stringified before sending to Mailgun. Note there's no need to include the code variable since it will be automatically added by the connector.

Example

The following is an example of the deliveries config:

{
  "Register": {
    "subject": "{{code}} is your verification code",
    "replyTo": "Foo <[email protected]>",
    "html": "<h1>Welcome to Logto</h1><p>Your verification code is {{code}}.</p>",
    "text": "Welcome to Logto. Your verification code is {{code}}."
  },
  "SignIn": {
    "subject": "Welcome back to Logto",
    "replyTo": "Foo <[email protected]>",
    "template": "logto-sign-in",
    "variables": {
      "bar": "baz"
    }
  }
}

Save your configuration

Double check you have filled out necessary values in the Logto connector configuration area. Click "Save and Done" (or "Save changes") and the Mailgun connector should be available now.

Once you create a connector successfully, you can enable phone number-based passwordless login and registration.

Navigate to Console > Sign-in experience > Sign-up and sign-in.
Set up sign-up methods (Optional):
1. Select "Email address" or "Email or phone number" as the sign-up identifier.
2. "Verify at sign-up" is forced to be enabled. You can also enable "Create a password" on registration.
Set up sign-in methods:
1. Select Email address as one of sign-in identifiers. You can provide multiple available identifiers (email, phone number, and username).
2. Select "Verification code" and / or "Password" as the authentication factor.
Click "Save changes" and test it in "Live preview".

In addition to registration and login via OTPs, you can also have password recovery and -based security verification enabled, as well as linking Email address to profile. See End-user flows for more details.

Testing and Validation

Return to your Chrome extension app. You should now be able to sign in with Mailgun. Enjoy!

Create an application in Logto​

Integrate Chrome extension with Logto​

Installation​

The authentication flow​

Regarding redirect-based sign-in​

Update the manifest.json​

Set up a background script (service worker)​

Update Logto application settings​

Add sign-in and sign-out buttons to the popup​

Checkpoint: Test the authentication flow​

Check authentication state​

Other considerations​

Add Mailgun connector​

Set up Mailgun email connector​

Prerequisites​

Basic configuration​

Deliveries​

Config object​

Usage types​

Content config​

Example​

Save your configuration​

Enable Mailgun connector in Sign-in Experience​

Testing and Validation​

Further readings​

Create an application in Logto

Integrate Chrome extension with Logto

Installation

The authentication flow

Regarding redirect-based sign-in

Update the `manifest.json`

Set up a background script (service worker)

Update Logto application settings

Add sign-in and sign-out buttons to the popup

Checkpoint: Test the authentication flow

Check authentication state

Other considerations

Add Mailgun connector

Set up Mailgun email connector

Prerequisites

Basic configuration

Deliveries

Config object

Usage types

Content config

Example

Save your configuration

Enable Mailgun connector in Sign-in Experience

Testing and Validation

Further readings