Using UnifyID GaitAuth™ with IFTTT Rules

Security for your digital life is more important today than ever, but it can be a real pain. Mobile authentication typically means getting out your phone, unlocking it, opening an app, remembering a password, authenticating in the app. Authentication via biometrics like fingerprints or facial recognition are a step in the right direction, but assume you have an ungloved finger or unmasked face available, among other potential shortcomings. 

UnifyID offers a novel new method of uniquely authenticating using the way you walk. With GaitAuth, an app can provide authenticated user functionality as long as the user is carrying their phone as they walk around. 

In this article, we’ll show how easy it is to integrate GaitAuth into a mobile app and use GaitAuth to authenticate the device’s user before triggering an IFTTT Webhook that could be used for sensitive tasks like home automation and security.

What is GaitAuth?

GaitAuth aims to remove the friction from mobile authentication. 

Anyone who has used a modern mobile device knows that passwords and personally identifiable information are not convenient nor, in practice, all that secure. Multi-factor authentication can help, but many popular authentication factors have drawbacks. For example, codes sent via SMS are susceptible to attacks like SMS spoofing and SIM hijacking.

GaitAuth uses motion-based behavioral biometrics and environmental factors to create a unique digital fingerprint of a mobile device’s user. GaitAuth uses this fingerprint to ensure the user is who they claim to be. This kind of passive authentication means no user disruption and real-time protection.

Even better, the GaitAuth SDK lets you easily incorporate GaitAuth into your mobile apps to provide machine learning-powered authentication. 

How does IFTTT fit in?

IFTTT is a popular automation platform that enables even non-technical users to create automations. Using IFTTT, it’s easy to set up workflows that perform actions in one or more services based on events triggered by a device or service. 

For example, a user with a Ring doorbell could use IFTTT to turn on their house’s interior smart lights if the doorbell detects motion in front of the house. But when manipulating user data and working with home automation devices like smart locks, it’s important to verify the user’s identity. 

For example, what if you could unlock a smart lock for your child just by having them walk up to the door? You want to be certain it’s your child walking up to the door before unlocking it. GaitAuth can provide that certainty. 

Getting started with GaitAuth and iOS

Let’s take a look at how you might use GaitAuth to authenticate IFTTT automations in an iOS app. Before you start, make sure you have CocoaPods installed. It’s the preferred installation method for both the GaitAuth and IFTTT iOS SDKs. 

You’ll also need to sign up for a UnifyID developer account so you can obtain an SDK key.

Start by opening Xcode and creating a new iOS app. To keep things straightforward, create a single-view iOS app using Storyboards.

Next, set up a Podfile in your app’s Xcode project directory, and follow the instructions for installing the GaitAuth and IFTTT dependencies. You’ll end up with a Podfile like this:

target 'GaitAuthIFTTT' do
    pod 'UnifyID/GaitAuth'
    pod 'IFTTTConnectSDK'
end

# Enable library evolution support on all dependent projects.
post_install do |pi|
    pi.pods_project.targets.each do |t|
        t.build_configurations.each do |config|
          config.build_settings['BUILD_LIBRARY_FOR_DISTRIBUTION'] = 'YES'
        end
    end
end

Run pod install from the terminal to install both SDKs. Now add the code needed to set up and use GaitAuth. Start by adding this line to the top of AppDelegate.swift:

import UnifyID

Then, initialize UnifyID just inside the AppDelegate class:

let unifyid : UnifyID = { try! UnifyID(
    sdkKey: "https://xxx@config.unify.id",
    user: "unique-immutable-user-identifier"
)}()

You can generate a real SDK key in the UnifyID developer portal. The user attribute can be set to anything you’d like, as long as no two users of your app have the same identifier. Once you’ve chosen an identifier for a user, use the same value every time that person uses your app.

Using GaitAuth in your iOS apps

The first step in using GaitAuth is creating and training a model. When training is complete, GaitAuth uses this model to identify and authenticate the user of the device. 

To train a GaitAuth model, we’ll use the following steps:

  1. Create a model on an iOS device.
  2. Add features to the model. Features represent data about the way the iOS device’s user walks. We’ll need to gather this data to train a GaitAuth model.
  3. Enqueue the model for server-side training.
  4. Check the status of the model on the server until it is ready.
  5. Download the trained model from the server to the device where your app is running.
  6. Use the trained model in your app to score newly collected features — which GaitAuth does automatically as the user walks around while carrying the device. This lets you authenticate quickly and easily.

Let’s examine how this process looks in Swift. To start using GaitAuth in your app, obtain an instance of it by calling:

let gaitAuth = unifyid.gaitAuth

Before we can effectively use GaitAuth in an iOS app, we’ll have to train a GaitAuth model to learn about your app user’s gait. Next, we create a model:

gaitAuth.createModel { result in
    switch result {
    case .success(let gaitModel):
        // Save gaitModel.id
    case .failure(let error):
        // Handle the error
    }
}

It’s important to save the model ID so we’ll be able to re-load this model from the server if necessary. With the model created, we can start gathering data about the device user’s gait:

gaitAuth.startFeatureUpdates(to: DispatchQueue.main) { result in
    switch result {
    case .success(let features):
        // Called when feature collection is complete
    case .failure(let error):
        // Handle the error
    }
}

Seven days’ training time is optimal. When the app has gathered enough feature data, call:

gaitAuth.stopFeatureUpdates()

This will call the .success result handler, and the features will be ready to use. Note that you need to use an iOS background mode to ensure model training continues even when the app isn’t currently on-screen. If this isn’t possible, use the feature serialization functionality described in the GaitAuth documentation to save the feature data that’s been gathered. This lets the app add to the existing feature collection when execution resumes. 

Next, we add features to the model:

gaitModel.add(features) { error in
    if let error = error {
        // Handle the error
        return
    }
    // Successfully added gait features to model
}

…and start the training process:

gaitModel.train() { error in
    if let error = error {
        // Handle the error
        return
    }
    // Training is in progress. Notify the user if necessary.
}

Training occurs on the GaitAuth server, and can take a few hours. When training is complete, the model’s .status property will be ‘ready‘, so the app should periodically check the status to determine when the model is ready to use for authentication.

Setting up an IFTTT Webhook

Now we’ll set up an IFTTT Webhook. Before proceeding, create a free IFTTT account. 

Start by opening the IFTTT dashboard at https://ifttt.com/home. Then, click ‘Create‘ to add a new Applet:

A picture containing text

Description automatically generated

Then, add an ‘If This’:

A picture containing text, clipart

Description automatically generated

Search for Webhooks and select it:

Graphical user interface, application, Teams

Description automatically generated

And choose ‘Receive a web request’:

Graphical user interface, text, application

Description automatically generated

Enter an event name like ‘gaitauth_trigger‘, and click ‘Create Trigger‘. Finally, add a ‘Then That’ to determine what happens when the webhook is called:

A picture containing icon

Description automatically generated

The task chosen here depends on what action we want the GaitAuth-enabled app to trigger. For example, if the device user has a smart lock and wants to unlock their door once they’ve been authenticated by GaitAuth, they can set up a ‘Then That’ to do exactly that. 

This would be particularly useful in a scenario where a parent would like to automatically unlock a house door when their child is coming home from school and approaches the house. There would be no need to remember to bring a key, and there would be no chance of getting locked out. The app, running in the background, can determine when the child is near home, use GaitAuth to verify that the device is being carried by the right person, and call the IFTTT Webhook to unlock the door.

The included sample app triggers an IFTTT Webhook when the device enters a geofenced area surrounding a user-provided address, but it only does so if GaitAuth authentication is successful. It can be used to perform any action that IFTTT is able to trigger – including unlocking a door when a child is nearly home as described above.

Next Steps

Congratulations, you’ve completed our high-level look at how to use GaitAuth and IFTTT together in an iOS app, with a few deeper dives into code at key points. To see what you’ve learned in action, we’ve created a complete sample app that you can find at https://github.com/UnifyID/blog-ios-ifttt

This is just the beginning! While GaitAuth and IFTTT make a great team, just imagine all the places where your iOS apps could benefit from extremely accurate real time authentication based on behavioral biometrics. 

Sign up for a UnifyID developer account at https://developer.unify.id/ and start building ML-powered gait authentication into your apps today.

Unlocking a Password Vault Based on Who You Are, Not What You Have

Security for your digital life is more important than ever. Passwords have been common up until now, but identity and authentication methods are moving away from usernames and passwords. They require users to sacrifice convenience for security and add friction to the overall experience — and many people choose convenience over security due to that friction. 

For instance, if the same password is used for more than one system, then the password being compromised on one system could make the user vulnerable on other systems. 

Password managers, vaults, and multi-factor authentication (MFA) can be useful tools, but much of the time we’re just adding another layer of passwords to remember. Sending authentication codes via SMS or email has been used as an element of MFA processes, but it relies on “what you have” — a device, an email address, or a phone number — and is subject to spoofing or redirection attacks. 

What if we could be even more specific about authentication? Not just what you have, but who you are?

Unique attributes of each person include various aspects of their biometrics, such as their fingerprints or face. While unique, these identifiers might not be available in environments where the user must wear a mask or gloves. Not all devices have biometric support. 

Another way to ensure a person is who they claim to be is by making use of their behavioral biometrics and environmental factors. This type of authentication is passive, or implicit, meaning a user can be authenticated without disruption. It’s also continuous. As a user moves around, the application is able to detect from one moment to the next whether the person carrying the device is the intended user. 

UnifyID offers a novel method of uniquely authenticating using the way you walk. With GaitAuth, your application can provide authentication just by the user carrying a phone as they walk around. In this article, we will show how easy it is to integrate GaitAuth into an Android application.

GaitAuth in an Android App

The GaitAuth SDK lets you easily incorporate machine learning-powered gait authentication into your applications. 

For this article, I created an example of a vault application that stores secrets intended to be accessed by only one user. Since we want to focus on the ease of integrating GaitAuth, the sample just stores simple text secrets like passwords, key strings, or other confidential data. But it could be easily extended to provide one-time passwords (OTP) using a Time-based One-time Password (TOTP) or HMAC-based One-time Password  (HOTP) algorithm. 

The application presents someone with a password challenge, and once the app authenticates the user from either gait data or a correct password, it displays the secrets.

Before integrating GaithAuth into an application, you will need to sign up for a free developer account. Once you have created an account, you will have access to the Developer Dashboard. 

On the Dashboard, next to SDK Keys, click Create. You’ll be asked for a name for identifying the key. Enter the name of your application or some other meaningful label here. After a name is entered and saved, the SDK key will display on the dashboard. You’ll need this key in your mobile app for SDK initialization. For the sample code accompanying this article, place your key in the Android resource file values/secrets.xml into the item named sdkKey.

To add the SDK to the application, a few lines must be added to the project and application build.gradle file. For the project build.gradle, mavenCentral must be added to the buildscript and allprojects sections.

buildscript {
   repositories {
       google()
       jcenter()
      mavenCentral() // add this line
   }
}

allprojects {
   repositories {
       google()
       jcenter()
      mavenCentral() // add this line
   }
}

In the build.gradle for the module, a line must be added to the dependencies section.

dependencies {
   implementation fileTree(dir: "libs", include: ["*.jar"])
   implementation 'androidx.appcompat:appcompat:1.2.0'
   implementation 'androidx.constraintlayout:constraintlayout:2.0.1'
   implementation 'id.unify.sdk:sdk-gaitauth:1.3.13' // add this line
   testImplementation 'junit:junit:4.12'
   androidTestImplementation 'androidx.test.ext:junit:1.1.2'
   androidTestImplementation 'androidx.test.espresso:espresso-core:3.3.0'
}

With these changes, a reference to the UnifyID GaitAuth SDK has been added. 

Some permissions are needed that the application might not already have. Within the application’s AndroidManifest.xml file, add the following permissions. 

<uses-permission android:name="android.permission.ACCESS_WIFI_STATE" />
<uses-permission android:name="android.permission.INTERNET" />
<uses-permission android:name="android.permission.ACCESS_NETWORK_STATE" />

GaitAuth learns how to recognize a user by collecting features from how the person walks. The user should carry the device around for a week so that your application can get to know how the person walks in various situations that may be part of their weekly routine. 

After collecting this feature data, your application will use it to train an ML model for recognizing the user’s steps. A trained model provides a confidence score for the current user being the intended user or for being an imposter. 

Setting Up Feature Collection

The first modification that we want to make to the application is for it to collect features as the user is walking. For collecting features, we place code within an Android service. The service can continue to run and collect features even if the user has navigated to a different application on the device. 

When the service starts, it will initialize the GaitAuth SDK, show a notification to let the user know that it is running, and begin collecting features. 

@Override
public int onStartCommand(Intent intent, int flags, int startId) {
    if(!isInitialized) {
        super.onStartCommand(intent, flags, startId);
        // The username is passed from the activity that starts the service
        userID = intent.getStringExtra(INTENT_USER_ID);
        initGait();
        initNotification();
        isInitialized = true;
    }
    return START_NOT_STICKY;
}

void initGait() {
   UnifyID.initialize(getApplicationContext(), GetSDKKey(), userID, new CompletionHandler() {
    @Override
    public void onCompletion(UnifyIDConfig config) {
        GaitAuth.initialize(getApplicationContext(), config);
        initModel();
    }
    @Override
    public void onFailure(UnifyIDException e) {
        e.printStackTrace();
        postUpdate(e.getMessage());
        Log.e(TAG, e.getMessage());
    }
    });
}

The gait model object is used for recognizing a user by the features of their steps. When a gait model is created, it receives a unique ID string. The ID string should be saved. The ID string can be used to reload the model when the application restarts later. 

void initModel() {
   // If we have not already instantiated a GaitModel, then create one.
    if(gaitModel == null ) {
        // See if there is a GaitModel ID that we can load.
        SharedPreferences pref = getSharedPreferences(
            PREFERENCE_GAITMODEL, MODE_PRIVATE);
        String modelID = pref.getString(PREFERENCE_KEY_MODELID,"");
        GaitAuth gaitAuth = GaitAuth.getInstance();
        try {
            // If there is no modelID, then create a model and save its ID
            if (modelID == "") {
                gaitModel = gaitAuth.createModel();
                SharedPreferences.Editor editor = pref.edit();
                editor.putString(PREFERENCE_KEY_MODELID, gaitModel.getId());
                editor.commit();
            } else {
                // If there is a modelID, then use it to load the model
                gaitModel = gaitAuth.loadModel(modelID);
            }
        } catch (GaitModelException exc) {
            exc.printStackTrace();
       }
    }
}

Training the Model

Since the model hasn’t been trained, it is not able to recognize the user yet. Let’s collect some features to train the model. As new features are generated, they are added to a collection. When the collection reaches a prescribed size, it is saved to device storage. 

final int FEATURES_TO_HOLD = 250;
Vector<GaitFeature> gaitFeatureList = new Vector<GaitFeature>();

void startFeatureCollection() {
    try {
        GaitAuth.getInstance().registerListener(new FeatureEventListener() {
            @Override
            public void onNewFeature(GaitFeature feature) {
                gaitFeatureList.add(feature);
                if(gaitFeatureList.size()>=FEATURES_TO_HOLD) {
                    saveFeatures();
                }
            }
        });
    } catch (GaitAuthException e) {
        e.printStackTrace();
    }
}

The SDK provides methods for serializing and deserializing feature lists to byte data. We will use the static method GaitAuth.serializeFeatures to make a byte stream from the feature list. This byte stream is then written to storage. 

void saveFeatures()  {
    if (gaitFeatureList.size() == 0) {
        return;
    }
    Vector<GaitFeature> featuresToSave = new Vector<GaitFeature>();
    synchronized (gaitFeatureList) {
       featuresToSave.addAll(gaitFeatureList);
       gaitFeatureList.clear();
    }
    try {
        byte[] featureData = GaitAuth.serializeFeatures(featuresToSave);
        File storageFile = getStorageFile(getNextFileSegment());
        FileOutputStream fos = new FileOutputStream(storageFile);
        fos.write(featureData, 0, featureData.length);
        fos.close();
        addFeatureCount(featuresToSave.size());
        notificationBuilder.setContentText(String.format("Saved feature set %d containing %d elements at %s", getFeatureCount(), featuresToSave.size(), new Date()));
        NotificationManager manager = getSystemService(NotificationManager.class);
        manager.notify(1, notificationBuilder.build());
    } catch (FeatureCollectionException | FileNotFoundException exc) {
        exc.printStackTrace();
    } catch (IOException exc) {
        exc.printStackTrace();
    }
}

Once there are enough features, we can initiate training. UnifyID recommends a minimum of three days and 7,000 walk cycles, but suggests using seven days and 10,000 walk cycles for optimal training.

In the sample program, a button on the settings screen starts the training. 

To train the model, we deserialize the features that have been collected and input them into our model using the add method. After the features are added, we call GaitModel.train. Training can take a few hours. GaitModel.getStatus gets the status of the model. The function returns one of the following values:

  • CREATED – the model hasn’t yet been trained.
  • TRAINING – training is in process. Check again later for the training result.
  • READY – the model is trained and ready to begin recognizing users
  • FAILED – the model could not be trained. 

If training fails, GaitModel.getReason returns a string that explains why the failure occurred. Attempting to train the model with a tiny data set results in a message stating that the amount of training data is insufficient. When the function returns READY, we can begin using it to authenticate the user. 

Authenticating the User

For a model that is ready, there are two methods of evaluating this data to determine if a device is in the hands of the intended person. 

One method is to examine the gait feature score. Possible scores range from -1.0 to 1.0. 

Positive scores indicate the current user is the person whose walking patterns were used to train the model. 

Negative scores indicate the person holding the device is an imposter. 

Your organization may want to evaluate different thresholds for acceptance to find one that is acceptable. A passing score of 0.6 to 0.8 is a good starting point. 

In the following code, the scores of the last four features collected are averaged together. If the most recent feature was collected within the past 60 seconds and the average is greater than 0.6, then the user is considered authenticated.

static final float PASSING_SCORE = 0.6f;
static final long  MAX_PASSING_AGE = 30000;
static final int MIN_SCORE_COUNT = 4;

public boolean isAuthenticated() {
    if(gaitScoreList.size()>=MIN_SCORE_COUNT) {
        long age = (new Date()).getTime() - gaitFeatureList.get(gaitFeatureList.size()-1).getEndTimeStamp().getTime();
        float sum = 0.0f;
        for(GaitScore score:gaitScoreList) {
            sum += score.getScore();
        }
        float avg = sum / (float)gaitScoreList.size();
        if(avg > PASSING_SCORE && age <=MAX_PASSING_AGE) {
            return true;
        }
    }
    return false;
}

When examining the feature scores to authenticate the user, there is freedom for deciding if a user will be authenticated or not for certain scenarios. 

There is also a much easier way to authenticate a user. The GaitAuth SDK provides an authenticator that can collect features and perform authentication for you. 

To create a GaitAuth authenticator, first create a configuration. The configuration contains settings for the maximum age of features to consider, setting a threshold for a passing feature, and other attributes that affect how the features are evaluated.

GaitQuantileConfig config = new GaitQuantileConfig(QUANTILE_THRESHOLD);
config.setMinNumScores(1);    // Require at least 1 score
config.setMaxNumScores(50);   // Set the maximum number of scores to use for authentication
config.setMaxScoreAge(10000); // Set the maximum age, in milliseconds, for features
config.setNumQuantiles(100);  // Set the number of quantiles (divisions) for the feature data
config.setQuantile(50);

Once created, the authenticator continues to collect information from the user’s walking. The authentication status can be checked at any time by calling getStatus() on the authenticator. The call to getStatus() accepts an  AuthenticationListener. Either the onCompletion or onFailure method on this object will be called. 

Note that if onCompletion is called, that does not imply that the user was authenticated. A call to onCompletion means that the authenticator was able to perform an evaluation. To determine if the user is authentic, check isAuthenticated.

gaitAuthenticator.getStatus(new AuthenticationListener() {
   @Override
   public void onComplete(AuthenticationResult result) {
       gaitAuthenticatorResult = result.getStatus();
              
       switch(result.getStatus())
       {
           case AUTHENTICATED:
               isAuthenticated = true;
           break;
           case UNAUTHENTICATED:
               isAuthenticated = false;
           Default:
	     break;
       }
   }
   @Override
   public void onFailure(GaitAuthException cause) {

   }
});

The authenticator will continue to authenticate the user in the background. 

The application unlocks the stored secrets when it detects the user recently walked and that it was an authorized user. If the application cannot authenticate the user from their walk (the user hasn’t recently walked), the password unlock feature is available as a fallback.

Next Steps

We now have a password vault application that is able to recognize a user by the way that they walk. We did this by adding the GaitAuth SDK to the project, collecting features based on the user’s gait, and using the features to train a model to recognize the user. The application unlocks for the user when the trained model recognizes their gait. As noted earlier, we can easily change the application functionality to add features like one-time passwords.

You can read more about the GaitAuth SDK at UnifyID’s documentation site. Sign up for free on the developer dashboard to begin testing with the SDK. For a more in-depth look at how gait biometric verification works, see this publication on the UnifyID site. You can download the sample application used in this blog post from https://github.com/UnifyID/blog-android-secrets.

Integrating Trusted Registration for PushAuth™

Welcome to the final post of the Power of PushAuth™ blog series. In this post, we will enhance the basic website we created in the Building a Web Application with PushAuth™ post by integrating trusted registration to provide a way for the mobile SDK to only register authorized users.

For a detailed explanation on trusted registration, please refer to the following:

In this tutorial, we will integrate trusted registration into a web application so that a user, upon signup, is given a 4 digit “pairing code” they can enter in the mobile app. After this, the mobile client added will be the only one who is able to receive login request push notifications.

The end result of this blog post will be almost identical to the publicly available sample web application used in the previous blog post that introduced trusted registration.

Setup

To follow this tutorial, you will need:

This tutorial assumes a basic familiarity with the Rails framework. Also, the starting point of this guide is the Rails app we built in the Building a Web Application with PushAuth™ post. If you have not followed this previous post to build the web application from scratch, you can clone the pushauth-sample-server in our GitHub repository as the starting point. If you choose to use the GitHub project as a starting point, make sure to initialize the project by following the steps in README of the project.

Step 1 – Provide a pairing code for users

Database Setup

First, we need to add new data to the User table in the database to keep track of an integer pairing code and a boolean which tracks whether or not the code has been used.

Let’s generate a migration to add these columns:

$ rails generate migration add_verification_code_to_users verification_code:integer

The newly generated migration should look like this:

# db/migrate/{some_date}_add_verification_code_to_users.rb

class AddVerificationCodeToUsers < ActiveRecord::Migration[6.0]
  def change
    add_column :users, :verification_code, :integer
  end
end

We also want to add a column to keep track of whether each pairing code has been used. This column should default to false so that we don’t have to set it when we create a user. Add this line under the other add_column line from the migration file:

add_column :users, :verification_used, :boolean, default: false

Run bundle exec rails db:migrate, and we should have the appropriate database setup.

User Model

Let’s add some business logic to our User model so that we can ensure the consistency of our table.

We want to add validation to the username such that it must be unique, and we want to add a before_create hook that generates a random verification code for a new User. While we’re at it, we may as well make a function that tries to use up a validation code, and returns whether or not it was a success.

# app/models/user.rb

class User < ApplicationRecord
  has_secure_password

 validates :username, presence: true, uniqueness: { case_sensitive: false }

 before_create :generate_verification_code

 def consume_verification_code(provided_code)
   if provided_code == self.verification_code && !self.verification_used
     self.update_attributes(verification_used: true)
   end
 end

 private

 def generate_verification_code
   self.verification_code = SecureRandom.rand(1000...9999)
 end
end

A quick note about consume_verification_code: in absence of any statements following the if, Ruby will return a boolean corresponding to the evaluated condition, so this will return true if and only if the code was correct and had not been used before.

Users Controller and View

Now, let’s make a UsersController so that we can implement user registration:

$ rails generate controller Users

In UsersController, we will add three actions

  • new : a signup form
  • create : the endpoint to which the signup data is sent
  • post_signup : the page displaying the verification code to the user
# app/controllers/users_controller.rb

class UsersController < ApplicationController
  skip_before_action :authorized

  def new
    @user = User.new
  end

  def create
    @user = User.new(params.require(:user).permit(:username, :password))

    if @user.save
      session[:signup_username] = @user.username
      session[:signup_verification_code] = @user.verification_code
      redirect_to post_signup_users_path
    else
      render :new
    end
  end

  def post_signup
    @username = session[:signup_username]
    @pairing_code = session[:signup_verification_code]
  end
end

Now, let’s add some new views.

app/views/users/new.html.erb for the signup page:

# app/views/users/new.html.erb

<%= form_for @user, class: "form-signin" do |f| %>
<% if @user.errors.any? %>
    <div class="error_messages">
      <h2>Form is invalid</h2>
      <ul>
        <% @user.errors.full_messages.each do |message| %>
          <li><%= message %></li>
        <% end %>
      </ul>
    </div>
  <% end %>
  <%= f.label :username, class: "sr-only" %>
  <%= f.text_field :username, class: "form-control", placeholder: "Username", autofocus: true %>

  <%= f.label :password, class: "sr-only" %>
  <%= f.password_field :password, class: "form-control", placeholder: "Password" %>

  <%= f.submit "Sign up!", {class: ["btn", "btn-lg", "btn-primary", "btn-block"]} %>
<% end %>

app/views/users/post_signup.html.erb displays the pairing code.

# app/views/users/post_signup.html.erb

<p>
Thanks for registering! Please enter the following in our app:
</p>

<pre>
  Username: <%= @username %>
  Pairing code: <%= @pairing_code %>
</pre>

And finally, we add our routes:

# config/routes.rb
# somewhere within the main do...end block

  resource :users, only: [:new, :create] do
    get "post_signup"
  end

We now want to add signup links from a couple different pages:

Replace the Welcome! Please <%= link_to "log in", "login" %>. line in app/views/application/home.html.erb with the following:

Welcome! Please <%= link_to "log in", "login" %> or <%= link_to "sign up", new_users_path %>.

And add this to the bottom of app/views/sessions/new.html.erb:

Don't have an account yet? Sign up <%= link_to "here", new_users_path %>!

At this point, you should be able to test the new signup flow by running bundle exec rails server.

Step 2 – Implement Trusted Registration Webhook Endpoint

When the user enters their name and pairing code in the app, the UnifyID PushAuth™ service will send a POST request to the Rails app for permission to add a device. On the Developer Dashboard, you can set up the endpoint URL. A detailed explanation on how the trusted registration webhook endpoint works can be found in the previous blog post.

We will make an endpoint for user verification at /users/trust, which checks if a given user and a challenge token (i.e., the pairing code for this sample website) match with our database records.

# app/controllers/users_controller.rb
# add these lines right under "class UsersController < ApplicationController"

  skip_before_action :authorized
  skip_before_action :verify_authenticity_token, only: :trust
  http_basic_authenticate_with name: Rails.application.credentials.unifyid[:basic_username],
    password: Rails.application.credentials.unifyid[:basic_password],
    only: :trust
    
  def trust
    @user = User.find_by(username: params[:user])
    if @user && @user.consume_verification_code(params[:challenge].to_i)
      head :ok
    else
      head :unauthorized
    end
  end 

Let’s take a look at these in a bit more detail:

  skip_before_action :verify_authenticity_token, only: :trust

By default, Rails has cross-site request forgery protection for forms, which means that a form that submits data via POST will have an authenticity token that makes it difficult for attackers to manipulate your browser into performing unauthorized requests. You can read more about this before_action here.
For our use case, however, the UnifyID service will not know the proper authenticity token, so we relax that requirement on the trust action.

  http_basic_authenticate_with name: Rails.application.credentials.unifyid[:basic_username],
    password: Rails.application.credentials.unifyid[:basic_password],
    only: :trust

This adds HTTP Basic Authentication to the /user/trust endpoint in order to make sure the request is from the UnifyID PushAuth™ service. Note that you will have to add basic_username and basic_password entries to the Rails application credentials (bundle exec rails credentials:edit).

Finally, we can add our route:

# config/routes.rb

  resource :users, only: [:new, :create] do
    post "trust"

    get "post_signup"
  end

And there you have it! We’ve added user signup, and we have secured mobile client registration to to UnifyID PushAuth™ service. The end result should look like our pushauth-sample-server-reg project in GitHub, which was introduced in the Power of Trusted Registration blog post.

This concludes the Power of PushAuth™ blog series. Thanks for following along, and feel free to reach out to us if you have any questions, comments, or suggestions!

The Power of Trusted Device Registration

Welcome once again to the Power of PushAuth™ blog series! This post builds on the previous posts by providing an extension of our sample project with feature and security enhancements.

The Problem of Unchecked Device Registration

In the “Is push authentication perfect?” section of the first post in the series, we pointed out some shortcomings of relying on push authentication as a means to authenticate users. Recall the following from that section:

“There’s also the issue of trusting a device to be associated with its true user during registration, as well as not allowing attackers to register devices under the true user’s account.”

The first iteration of the open source sample project did not provide a solution to this problem. With that initial, simple implementation of PushAuth, there was virtually no confidence in the devices registered under a given username. If an attacker were able to obtain the SDK key and one or more usernames registered on the server, they could configure the app on their phone and fraudulently accept or reject push notification login attempts. This security hole allows attackers to impersonate true users and access their resources in the website, or to lock a true user out of accessing their own resources. Not great, considering this is supposed to be a method of increasing the security of the authentication flow.

Trusted Device Registration

The extension presented in this post, however, does include trusted device registration! It also provides the added feature of user registration in the website. Remember that the initial project required users to be created in the console. Only users whose username and password were inserted to the database in that manner were able to log in to the website and receive push notifications. Now users register in the website directly. Once a user signs up in the website, they configure the sample app on their phone in order to authorize future login attempts. This is where trusted device registration comes into play.

What’s different for the user?

After dictating what username and password to associate with a new user, the website displays a four-digit pairing code. The user is instructed to enter that pairing code in the sample app with their chosen username. If the values match, then the user will receive push notifications during login attempts and be able to respond accordingly. If the username and pairing code do not match, then registration of that mobile device fails. This means they will not be able to receive push notifications to confirm or deny login attempts. An attacker will not know that unique pairing code, so they are unable to impersonate a user in the app.

How does this work?

The PushAuth sample server now has a verification endpoint, /users/trust, which accepts webhook requests from UnifyID’s PushAuth service. UnifyID’s server makes an HTTP POST request to a configured target URL, which is set in the project’s dashboard. The request body contains the username and the pairing code. The /users/trust endpoint looks up the given username and determines if the provided pairing code matches the server-generated code displayed during that user’s registration. If the request returns 200, then the user is considered verified and the device trusted. However, if the request returns anything else, the user is not considered verified and device registration fails.

The webhook target URL must use the https scheme and requests use the ‘Basic’ HTTP Authentication scheme, which can be used to validate requests. Further, the four-digit alphanumeric code generated by the server is single-use. These details collectively make it harder for an attacker to impersonate users. An attacker does not have access to the pairing code during initial sign up and they also cannot reuse the pairing code, thus ensuring that devices registered with users are trusted to belong to the true user.

Note: The way we chose to implement trusted device registration with a pairing code is by no means the only method of implementing trusted registration via webhook.

What are the limitations?

If you remember from our first post in the series, there will seemingly always be security holes in a login flow. This enhancement is no exception. For one, the pairing code is only displayed during user registration. What if the user accidentally closes out the tab and then has no way to pair their device? Or what if something happens with the device or app on their device and they have no way to reconfigure an app to be tied to their user? There’s also the concern of in-person over-the-shoulder access to the pairing code, where an attacker could see the four digits and enter the code on their phone before the true user is able to do so. However, just like before, there are solutions to those problems as well; they just aren’t included here. The extension we provide is certainly a big security improvement and patches one of the larger holes that existed in the initial sample project.

Developer Changes

Now that we’ve gone over the concept of trusted device registration and the value it adds to the PushAuth sample project, we’ll walk you through the actual changes from a developer’s perspective. The new repositories can be found here:

We have videos on the UnifyID YouTube channel to walk you through the full end-to-end flow of adding PushAuth to a website login flow with trusted registration.

Trusted Registration Webhook

First and foremost, you’ll need to enable trusted registration for your project in the dashboard. Reference the Setting up Trusted Registration section of our documentation to do so.

The server needs to be publicly available for the target URL to be used. If you want to follow along the tutorial without hosting the web server, we suggest using ngrok to expose the sample rails server running locally on your machine to the internet. The steps shown in this post will do that.

$ ngrok http 3000

Session Status                online
Account                       morganfrisby (Plan: Free)
Version                       2.3.35
Region                        United States (us)
Web Interface                 http://127.0.0.1:4040
Forwarding                    http://9c6b07ea6861.ngrok.io -> http://localhost:3000
Forwarding                    https://9c6b07ea6861.ngrok.io -> http://localhost:3000

The target URL is the https URL forwarding to your local server, with /users/trust appended. For example:

Once you add the target URL, trusted registration is enabled for your project. The username and password displayed are used in HTTP Basic Authentication. The username value is set to unifyid, and the password is an auto-generated random string. This value can be rotated, should it become compromised.

Now that the trusted registration webhook is set up in the dashboard, you’re ready to move on to setting up the server.

Server Setup

Other than that, server setup more or less follows the instructions from the first tutorial. Since users are now registered on the website, there is no need to create users in the console during server setup. The other change is the addition of basic_username and basic_password to the credentials file, which are the values found in your project dashboard under the Trusted Registration Webhook section, shown above.

The streamlined steps to spin up the server are now:

$ git clone https://github.com/UnifyID/pushauth-sample-server-reg.git
$ cd pushauth-sample-server-reg
$ bundle install
$ yarn install --check-files
$ bundle exec rails db:migrate

$ EDITOR=vim rails credentials:edit
unifyid:
  server_api_key: <your_key_goes_here>
  basic_username: <basic auth username for /users/trust webhook endpoint>
  basic_password: <basic auth password for /users/trust webhook endpoint>

$ bundle exec rails server

Feel free to reference the detailed tutorial in our previous post for more context around those commands, or the technical details blog post to understand how the server was built. Our next post in the series will provide the technical details of the trusted registration extension of the sample server.

You can see the new website screenshots here (notice the pairing code in the third screenshot):

Sample App Setup

Nothing changes from the developer’s perspective for the iOS and Android sample apps; simply clone the updated repositories and follow the same instructions from earlier in the blog series. Those two posts can be found here:

The iOS sample app screens will look like:

The Android sample app screens now look like:

That’s it! Thanks for following along with the Power of PushAuth™ blog series. The next, and final, post will include the technical details of adding trusted registration to the sample project. Until then, feel free to reach out to us with questions or comments at https://unify.id/contact-us/.

Building a Web Application with PushAuth™

Welcome back to the Power of PushAuth™ blog series! This is the fifth post of the Power of PushAuth™ blog series. The first post of the series was a comprehensive guide to push authentication. The subsequent three posts comprised an end-to-end sample implementation of PushAuth™ in a simple user login flow:

  1. Web Server tutorial
  2. iOS Mobile App tutorial
  3. Android Mobile App tutorial

In this post, we will create a sample website from scratch using Rails, and will integrate PushAuth™ APIs into the user login flow. Along the way, we will also provide technical details on how the website interacts with PushAuth™ APIs, which can help readers incorporate PushAuth™ into any existing website. The end result of this tutorial will be similar to the sample web server we deployed in Web Server tutorial.

Setup

To follow this tutorial, you will need:

This tutorial assumes a basic familiarity with the Rails framework.

Step 1: Make basic Rails app with simple session-based authentication

First, we will create website with a simple username/password authentication, without incorporating UnifyID PushAuth™. Step 2 will integrate PushAuth™ into the website we create in Step 1.

Project Initialization

$ rails new push_auth_demo
$ cd push_auth_demo

Let’s add the bcrypt gem to our Gemfile by uncommenting this line in Gemfile:

gem 'bcrypt', '~> 3.1.7'

Now, we will run bundle install to update the Gemfile.lock.

Next, we can add the User model to our database; each User will have a username and a password hash.

$ rails generate model user username:uniq password:digest 
$ rails db:migrate

Now, we can generate the controller for handling sessions.

$ rails generate controller sessions new create destroy

Controller Logic

The basic idea of session-based authentication is pretty simple:

  • When a user logs in, session[:user_id] is set to be the unique index of the corresponding User in the database.
  • When no one is logged in, session[:user_id] should be unset.

We’ll start with writing the Application controller, where we have a couple of simple page actions:

  • GET / (application#home) renders a page that shows links to other pages and actions
  • GET /restricted (application#restricted) renders a page that is only accessible when logged in.

A few helper functions will live here as well:

  • current_user should either return a User object, or nil if there’s no one logged in.
  • logged_in? should return whether someone is logged in
  • authorized is an action that is called before loading pages that require the a user to be logged in.
# app/controllers/application_controller.rb

class ApplicationController < ActionController::Base
  before_action :authorized
  helper_method :current_user
  helper_method :logged_in?

  skip_before_action :authorized, only: [:home]

  def current_user
    User.find_by(id: session[:user_id])
  end

  def logged_in?
    !current_user.nil?
  end

  def authorized
    unless logged_in?
      redirect_to login_path, alert: "You must be logged in to perform that action."
    end
  end
end

Now, we’ll handle users logging in or out through the Sessions controller.

  • GET /login (sessions#new) action displays a login page unless the user is already logged in
  • POST /login (sessions#create) will authenticate the user and set the session.
  • DELETE /logout (sessions#destroy) action will clear the user’s session cookie.
# app/controllers/sessions_controller.rb

class SessionsController < ApplicationController
  skip_before_action :authorized, except: [:destroy]

  def new
    redirect_to root_path if logged_in?
  end

  def create
    @user = User.find_by("lower(username) = ?", params[:username].downcase)
    if @user && @user.authenticate(params[:password])
      session[:user_id] = @user.id
      redirect_to root_path, notice: "Successfully logged in!"
    else
      redirect_to login_path, alert: "Sorry, that didn't work."
    end
  end

  def destroy
    session[:user_id] = nil
    redirect_to root_path, notice: "Successfully logged out."
  end
end

We also need to add routes for these actions.

# config/routes.rb

Rails.application.routes.draw do
  root "application#home"

  get "restricted", to: "application#restricted"

  get "login", to: "sessions#new"

  post "login", to: "sessions#create"

  delete "logout", to: "sessions#destroy"
end

Views

home page where you can click on the link to log in/out:

<!-- app/views/application/home.html.erb -->

<h1> Welcome! </h1>

<% if logged_in? %>
  Welcome, <%= current_user.username %>. <br />
  <%= link_to "Log out", logout_path, method: :delete %> <br />
  Click <%= link_to "here", restricted_path %> to see a super secret page!
<% else %>
  Please <%= link_to "log in", login_path %>.
<% end %>

restricted page to test access control:

<!-- app/views/application/restricted.html.erb -->

Shhh, this page is a secret!

sessions#new renders simple login form:

<!-- app/views/sessions/new.html.erb -->

<h1>Login</h1>

<%= form_tag "/login", {class: "form-signin"} do %>
  <%= label_tag :username, nil, class: "sr-only" %>
  <%= text_field_tag :username, nil, class: "form-control", placeholder: "Username", required: true, autofocus: true %>

  <%= label_tag :password, nil, class: "sr-only" %>
  <%= password_field_tag :password, nil, class: "form-control", placeholder: "Password", required: true%>

  <%= submit_tag "Log in", {class: ["btn", "btn-lg", "btn-primary", "btn-block"]} %>
<% end %>

Lastly, we’ll modify the default template to include a navigation bar at the top and flash messages for notice and alert from controllers:

<!-- Replace the contents of the <body> tag in app/views/layouts/application.html.erb with the following -->

    <nav class="navbar navbar-dark bg-dark">
      <a class="navbar-brand" href="/">
        <span class="logo d-inline-block align-top"></span>
        UnifyID PushAuth Sample
      </a>
    </nav>
    <% if flash[:notice] %>
      <div class="alert alert-primary alert-dismissible fade show" role="alert">
        <%= flash[:notice] %>
        <button type="button" class="close" data-dismiss="alert" aria-label="Close">
          <span aria-hidden="true">×</span>
        </button>
      </div>
    <% end %>
    <% if flash[:alert] %>
      <div class="alert alert-danger alert-dismissible fade show" role="alert">
        <%= flash[:alert] %>
        <button type="button" class="close" data-dismiss="alert" aria-label="Close">
          <span aria-hidden="true">×</span>
        </button>
      </div>
    <% end %>

    <main role="main" class="container">
      <div class="main-content">
        <%= yield %>
      </div>
    </main>

Styling

We can add styling to our website by simply adding bootstrap. The views we created above already use class names that are recognized by bootstrap.

First, add bootstrap and some of its dependencies.

$ yarn add bootstrap jquery popper.js

Then, add the following to the end of app/javascript/packs/application.js:

import 'bootstrap'
import 'stylesheets/application.scss'

Next, make a file called app/javascript/stylesheets/application.scss and add this:

@import "~bootstrap/scss/bootstrap";

Optionally, you may add your own custom CSS files as well. See an example in our sample code.

At this point, you should be able to run rails server and navigate to http://localhost:3000 to interact with the basic authentication server! You can create sample users as follows:

$ bundle exec rails console
> User.create(:username => "<your_username>", :password => "<your_password>").save
> exit

Step 2 – Integrate with UnifyID PushAuth™ APIs

Now that we have a website with a simple username/password authentication, let’s incorporate UnifyID PushAuth™ APIs to further enhance security.

Interface to PushAuth™ APIs

First, let’s tell Rails about your UnifyID API key.

Run rails credentials:edit and add the following

unifyid:
  server_api_key: <Your UnifyID API Key created from dashboard>

Next, add the following to config/application.rb, after the config.load_defaults line:

config.x.pushauth.base_uri = "https://api.unify.id"

Let’s also add the httparty gem to easily make HTTP/S requests. To do this, add the following to your Gemfile and run bundle install:

gem 'httparty', '~> 0.18.0'

Now, we will make a file called app/services/push_auth.rb which contains the interface for our Rails app to interact with the PushAuth™ APIs:

  • create_session method calls POST /v1/push/sessions to initiate PushAuth™ session (API doc).
  • get_session_status method calls GET /v1/push/sessions/{id} to retrieve the status of PushAuth™ session (API doc).
# app/services/push_auth.rb

class PushAuth
  include HTTParty
  base_uri Rails.configuration.x.pushauth.base_uri

  @@options = {
    headers: {
      "Content-Type": "application/json",
      "X-API-Key": Rails.application.credentials.unifyid[:server_api_key]
    }
  }

  def self.create_session(user_id, notification_title, notification_body)
    body = {
      "user" => user_id,
      "notification" => {
        "title" => notification_title,
        "body" => notification_body
      }
    }
    post("/v1/push/sessions", @@options.merge({body: body.to_json}))
  end

  def self.get_session_status(api_id)
    get("/v1/push/sessions/#{api_id}", @@options)
  end
end

Controller Logic Modification

Now, we will modify the login flow to incorporate PushAuth™ as the second factor authentication. The new login flow will consist of the following:

  1. The client submits the username and password via a POST request to /login (sessions#create)
  2. The controller validates that the user exists and the password matches. If not, it displays an error message.
  3. Upon successful username/password authentication, the controller creates a PushAuth™ session and redirects to GET /mfa (sessions#init_mfa)
  4. The Javascript in /mfa page repeatedly queries GET /mfa/check (sessions#check_mfa), which checks the PushAuth™ session status until the session status is no longer pending.
  5. Upon receiving a non-pending session status, the client submits a request to PATCH /mfa/finalize (sessions#finalize_mfa) that completes the login process.

First, let’s replace the create action in app/controllers/sessions_controller.rb:

  def create
    @user = User.find_by("lower(username) = ?", params[:username].downcase)
    if @user && @user.authenticate(params[:password])
      session[:pre_mfa_user_id] = @user.id

      pushauth_title = "Authenticate with #{Rails.application.class.module_parent.to_s}?"
      pushauth_body = "Login request from #{request.remote_ip}"

      response = PushAuth.create_session(@user.username, pushauth_title, pushauth_body)

      session[:pushauth_id] = response["id"]

      redirect_to mfa_path
    else
      redirect_to login_path, alert: "Sorry, that didn't work."
    end
  end

Next, let’s also add check_mfa and finalize_mfa actions in this controller:

  def check_mfa
    status = PushAuth.get_session_status(session[:pushauth_id])["status"]

    render plain: status
  end

  def finalize_mfa
    case PushAuth.get_session_status(session[:pushauth_id])["status"]
    when "accepted"
      session[:user_id] = session[:pre_mfa_user_id]
      session[:pushauth_id] = nil
      session[:pre_mfa_user_id] = nil
      flash.notice = "Successfully logged in!"
    when "rejected"
      session[:pre_mfa_user_id] = nil
      flash.alert = "Your request was denied."
    end
  end

We also want to make sure that only users who completed the password authentication are able to access actions for the PushAuth™ authentication. Thus, within sessions_controller we will add:

# app/controllers/sessions_controller.rb

# Add this just under the skip_before_action line
  before_action :semi_authorized!, only: [:init_mfa, :check_mfa, :finalize_mfa]

# And add this after action methods
  private
  def semi_authorized
    session[:pre_mfa_user_id] && session[:pushauth_id]
  end

  def unauthorized
    redirect_to login_path, alert: "You are not authorized to view this page."
  end

  def semi_authorized!
    unauthorized unless semi_authorized
  end

Views

Now, we need a page that uses AJAX to determine whether the PushAuth™ request has been completed.

First, let’s add a line in our application template that allows us to add content inside the <head> tag.
Add this to app/views/layouts/application.html.erb, right before the </head> tag:

<%= yield :head %>

Next, let’s add the Javascript code we want to run on the init_mfa page:

// app/javascript/packs/init_mfa.js

import Rails from "@rails/ujs";
let check_status = window.setInterval(function() {
  Rails.ajax({
    type: "GET",
    url: "/mfa/check",
    success: function(r) {
      if (r !== "sent") {
        Rails.ajax({
          type: "PATCH",
          url: "/mfa/finalize",
          success: function() {
            window.clearInterval(check_status);
            window.location.href = "/";
          },
          error: function() {
            console.log("Promoting PushAuth status failed.");
          }
        });
      }
    },
    error: function() {
      console.log("Checking for PushAuth status failed.");
    }
  });
}, 2000);

This will poll /mfa/check every 2 seconds, until the Rails app reports that the PushAuth™ request has been accepted, rejected, or expired. At this point, the browser will ask the Rails app to complete the login process by submitting a /mfa/finalize request.

Now, let’s add a view file for init_mfa that includes the Javascript above.

<!-- app/views/sessions/init_mfa.html.erb -->

<% content_for :head do %>
  <%= javascript_pack_tag 'init_mfa' %>
<% end %>

<div class="spinner-border" role="status" ></div>

Waiting for a response to the push notification...

Finally, we will add new mfa routes.

# Add these to config/routes.rb

  get "mfa", to: "sessions#init_mfa"

  get "mfa/check", to: "sessions#check_mfa"

  patch "mfa/finalize", to: "sessions#finalize_mfa"

Congratulations! We have now integrated UnifyID’s PushAuth™. The final result should function just like the pushauth-sample-server project, which we introduced in our How to Implement PushAuth™: Web Server post. Please reach out to us if you have any questions, comments or suggestions, and feel free to share this post.

How to Implement PushAuth™: Android Mobile App

This post is part of the Power of PushAuth™ blog series. The first post of the series was a comprehensive guide to push authentication. The next three posts of the series comprise an end-to-end sample implementation of PushAuth in a simple user login flow. The tutorial breakdown is as follows:

  1. Web Server tutorial
  2. iOS Mobile App tutorial
  3. Android Mobile App tutorial (this post)

The tutorial in this post builds on the web server from the first tutorial. With your web server set up and running, you now need a mobile app to receive and respond to notifications. This post will help you build the Android mobile app to do so; then you will be able to leverage the power of PushAuth for login requests!

Setup

To follow this tutorial, you will need:

  • An Android device running Android 7.0 or higher
  • A computer with Android Studio 4.0 installed
  • JDK 8 installed (we recommend using jabba)

Step 1 : Cloning the Project

The sample Android mobile app code for this project is provided in the pushauth-sample-app-android GitHub repository. Clone this repo to your local machine:

$ git clone https://github.com/UnifyID/pushauth-sample-app-android.git

Open Android Studio, select “Open an existing Android Studio project”, and select the directory of the cloned repository. It should look something like this:

Step 2: Set up Firebase Cloud Messaging (FCM)

Firebase Cloud Messaging (FCM) is the platform used to send notifications to Android devices. You will first need to set up Firebase in your app by following the instructions at https://firebase.google.com/docs/android/setup. After doing this, ensure that a google-services.json file is present under the app/ directory of the project in Android Studio.

Next, navigate to the Firebase project settings of your app in your browser and select the “Cloud Messaging” tab. The token labeled “Server key” is what you will need to provide to UnifyID in the next section, so copy that value.

Step 3: Providing Push Credentials to UnifyID

Now you have the Firebase Cloud Messaging (FCM) server key copied, you will provide it to UnifyID so that PushAuth can send push notifications to the sample app on your phone. You’ll provide this value in your project’s dashboard. Follow the instructions in the Developer Portal docs to do so. After you’ve done this, your project dashboard will indicate they are successfully uploaded:

Step 4: Building the Project

Now, back to Android Studio. Make sure that your Android device is plugged in to your computer, has developer tools and USB debugging enabled, and is available in the top center of Android Studio. Also make sure that the google-services.json file is located under the app/ directory of the project.

With everything set up, click the green triangle or press Control+R to run the app. The following screen should appear on your device:

Step 5: Mobile App Settings

You now have all the values necessary for configuration! Tap the gear icon in the top right corner of the sample app’s Configuration screen. For SDK key, enter your UnifyID project’s SDK key value from the Dashboard. The User string should be the same value that you used when creating a user in the web server tutorial, e.g. “Morgan”. If these values do not match, you will not be able to successfully respond to push notifications in the login flow.

After setting those values and clicking “Confirm”, the app is ready to receive your PushAuth login requests! The app will remain on the “Waiting for PushAuth requests” page until it receives a PushAuth authentication request.

Now you can go through the full login flow by entering your username and password on the login page, respond to the notification received by this app on your phone, and be successfully logged into the website.

That’s it! Now you have a simple login flow that integrates PushAuth. Stay tuned for the rest of the posts in the series, make sure to share this post and reach out to us if you have any questions, comments or suggestions!

How to Implement PushAuth™: iOS Mobile App

This post is part of the Power of PushAuth™ blog series. The first post of the series was a comprehensive guide to push authentication. The next three posts of the series comprise an end-to-end sample implementation of PushAuth in a simple user login flow. The tutorial breakdown is as follows:

  1. Web Server tutorial
  2. iOS Mobile App tutorial (this post)
  3. Android Mobile App tutorial

The tutorial in this post builds on the web server from the first tutorial. With your web server set up and running, you now need a mobile app to receive and respond to push notifications. This post will help you build the iOS mobile app to do so; then you will be able to leverage the power of PushAuth for login requests!

Setup

To follow this tutorial, you will need:

Step 1: Cloning the Project

The pushauth-sample-app-ios GitHub repository contains the sample iOS mobile app code for this project. Clone the repository to your local machine and open the PushAuthSample.xcworkspace file in Xcode.

$ git clone https://github.com/UnifyID/pushauth-sample-app-ios.git

Step 2: Setting Up and Running the Project

  1. In the top left section of your Xcode window, set the active scheme to PushAuthSample.
  2. Plug your phone into your computer. Your phone’s name will appear as the chosen device next to the active scheme.
  3. Navigate to the “Signing & Capabilities” section of the Xcode project settings.
  4. Check the boxes next to “Automatically manage signing” in the “Signing (Debug)” and “Signing (Release)” sections. This will simplify setup and merge the two into a single “Signing” section.
  5. Choose the “Team” value to match your Apple Developer account.
  6. Set the “Bundle Identifier” to something unique; this value will be used in the next step of the tutorial when you create the Identifier through the Apple Developer site.

After following these six steps, your settings should closely resemble the screenshot above from Xcode. Once everything is set up properly and with your phone still connected to your computer, run the project (Product > Run or Command-R). This screen will show up on your phone:

Step 3: Create an Apple Bundle Identifier

This step requires you to an Apple Developer Program Role with adequate permissions. The role-permissions are listed here.

Navigate to the Identifiers tab on the Certificates, Identifiers & Profiles page of the Apple Developer site. You’ll need to add a new identifier that matches the Bundle Identifier value you set in Xcode in step 6 above. Click the plus symbol next to the title at the top of the page; if you don’t see this symbol, you likely don’t have adequate permissions. Follow these instructions for the subsequent pages:

  1. Register a new identifier page: Keep the default selection (App IDs) and click “Continue”.
  2. Select a type page: Keep the default selection (App) and click “Continue”.
  3. Register an App ID page:
    • Description: enter an appropriate description for this project, e.g. “PushAuth Project”. This value will be displayed as the “Name” on the Identifiers page.
    • Bundle ID: Keep the selection on “Explicit” and enter the exact same value you put as the Bundle Identifier in the Xcode Signing & Capabilities page earlier.
    • Enable Push Notification capability by scrolling down on the page and selecting the checkbox next to “Push Notifications”.
    • Click “Continue”, verify everything was entered correctly, and click “Register”.

Now that you have created an identifier for this project, you can create a push notification certificate associated with this identifier.

Step 4: Create a Push Notification Certificate

UnifyID requires the APNs certificate in *.p12 format to send PushAuth requests to the app. This can be done from the same Identifiers page of the Apple Developer site that you were on in Step 3.

  1. Click on the name of the identifier you just created, e.g. “PushAuth Project”.
  2. Scroll down to the “Push Notifications” row and click on the “Configure” box. Next to this box you should see “Certificates (0)” since you haven’t yet created a certificate associated with this identifier.
  1. In the Apple Push Notification service SSL Certificates pop-up window, click on the “Create Certificate” box under “Production SSL Certificate” then click “Done”.
  1. At this point, you need to create a Certificate Signing Request (CSR) file from your Mac. Click “Learn More” and follow those instructions for doing so. Then upload that file and continue.
  1. Now that you have created a certificate, you must download it locally to export it to *.p12. Click “Download”.
  1. This will prompt you to add the certificate to Keychain Access. Choose a Keychain, e.g. “login”, to add the certificate to and click “Add”.
  1. Then find that certificate in Keychain Access. It may be useful to select the “Certificates” category and utilize the search bar to find the certificate you just added.
  1. Once you have located your certificate, right-click on it and click the option to export the certificate:
  1. Specify a name for the *.p12 file and a location to save it. Make sure the file format is set to “Personal Information Exchange (.p12)” then click “Save”.
  1. You will be prompted to password-protect the exported *.p12 file. Choose to export it without a password; simply click “OK”.

Now you have successfully created a APNs certificate in *.p12 format! This will be used by UnifyID and needs to be uploaded to your project settings through the dashboard.

Step 5: Providing Push Credentials to UnifyID

Now you have an Apple Bundle Identifier and an APNs push certificate. It’s time to provide your push credentials to UnifyID so that PushAuth can send push notifications to the sample app on your phone. Check out the Developer Portal docs here, or follow along the instructions below.

  1. Navigate to the “Push Credentials” section of your project on the Developer Dashboard.
  2. Click on “Choose File” and select the *.p12 file you generated in Step 4 of this tutorial.
  3. Choose the “Development/Sandbox APNs server” option for now since we are sending push notifications to an app that runs directly from Xcode. Later on, choose “Production APNs server” when you need to send PushAuth requests to apps distributed through the App Store or through ad-hoc means.
  4. Click “Add” to complete the upload.

Once the push credentials are successfully uploaded to your project settings, you will see the push credential information displayed:

If you find yourself needing to change the push credentials used for the project, simply click “Edit” and go through the same upload steps with the new credentials.

Step 6: Mobile App Settings

You now have all the values necessary for configuration! Open the sample app on your phone and tap the gear icon in the top right of the Configuration screen. For SDK key, enter your UnifyID project’s SDK key value from the Dashboard. The User string should be the same value that you used when creating a user in the web server tutorial, e.g. “Morgan”. If these values do not match, you will not be able to successfully respond to push notifications in the login flow.

Once you set those two values, you must allow push notifications for the app, then the app is ready to receive your PushAuth login requests!

Now you can go through the full login flow by entering your username and password on the login page, respond to the push notification received by this app on your phone, and be successfully logged in to the website.

That’s it! You now have a simple login flow that integrates PushAuth. The next post provides a tutorial for building the Android sample PushAuth mobile app. Stay tuned for the rests of the posts in the series and, as always, please share this post and reach out to us with questions, comments or suggestions.

How to Implement PushAuth™: Web Server

Welcome back to the Power of PushAuth™ blog series! The first post provided a comprehensive guide to push authentication — check it out here if you missed it. The next three posts are tutorials offering an end-to-end implementation of PushAuth™ in a simple user login flow and will be broken down as follows:

  1. Web Server tutorial (this post)
  2. iOS Mobile App tutorial
  3. Android Mobile App tutorial

The first tutorial (this post) covers a Ruby on Rails backend that provides a basic user login authentication flow with PushAuth integrated. The second and third tutorials will be instructions on how to run sample iOS and Android apps, respectively. These will be the apps that receive and respond to push notifications initiated by the login process.

By the end of these three tutorials, you will have a website where a registered user can log in with their username and password, receive a push notification on their phone, accept the login request via the push notification, and subsequently be logged in on the website. This flow is shown in the video below.

This is a very simplified version of a real-world application and login flow. You might remember some of the security issues that can be present with push authentication from the previous post, such as trusted device registration, fallback resources, or access revocation. This tutorial does not include solutions for those or user sign-up. Future posts in the series will provide extensions of this simple flow to tackle some of those issues.

Alright, let’s get started!

Setup

To follow this tutorial, you will need:

Step 1: UnifyID Account, Project, and Keys

A UnifyID project will grant you access to UnifyID’s services. In order to create a project, you’ll need to first create a UnifyID account. Once you have an account and project, go to the Developer Dashboard to create an API key and SDK key. Make sure to copy the API key value somewhere safe – you won’t be able to access it later.

This tutorial is using PushAuth project as the UnifyID project name. You can see the configured API and SDK keys on the dashboard view above.

Step 2: Cloning the Project and Installing Dependencies

The pushauth-sample-server GitHub repository contains the code for this project. Clone the repository, navigate into it, install the project dependencies that are listed in the Gemfile, and ensure your Yarn packages are up-to-date:

$ git clone https://github.com/UnifyID/pushauth-sample-server.git
$ cd pushauth-sample-server
$ bundle install
$ yarn install --check-files

Feel free to poke around the code if you’d like to get a better understanding of what’s going on under the hood. This tutorial won’t go into those details, but a future post in this series will.

Step 3: DB Setup and Running the Server

Now, initialize the database:

$ bundle exec rails db:migrate

Once the database has been initialized you can create users. To create a user, do the following (replacing <your_username> and <your_password> with the values you intend to use for username and password):

$ rails console
> User.create(:username => "<your_username>", :password => "<your_password>").save
> exit

Step 4: Server API Key Storage

This step requires the server API key value you copied in Step 1.

$ EDITOR=vim rails credentials:edit

The above command will open the credentials file in vim. You can replace vim with the name of whichever executable you are most comfortable (atom, sublime, etc.). This will decrypt and open the credentials file for editing, at which point you should add the following entry:

unifyid:
  server_api_key: <your_key_goes_here>

After saving and closing, the credentials file will be re-encrypted and your server API key value will be stored.

Step 5: Running the Server

Now you are able to run the server:

$ bundle exec rails server

Finally, connect to the server by opening http://localhost:3000/ in your browser. This will bring you to the landing page, where you can then navigate to the login page and enter your username and password from above in Step 3:

This brings you to the end of the web server tutorial. After entering the username and password on the login page, you can see that the server is polling for the push notification response before allowing or denying access to the website. Without a way to receive or respond to push notifications, you cannot successfully log in. Stay tuned for the next couple posts of this tutorial installment to complete the flow:

Thanks for following along! Please reach out to us if you have any questions, comments or suggestions, and feel free to share this post.

The Power of PushAuth™

Welcome to the first post in our series on PushAuth™, UnifyID’s push authentication service. This series, outlined below, will help you leverage push authentication to increase the security of your authorization flow.

PushAuth™ Blog Series Content

  1. Comprehensive guide to push authentication: what it is, how it works, and pros and cons of using it.
  2. An end-to-end tutorial of implementing PushAuth™ in a simple user login flow.
    1. How to Implement PushAuth™: Web Server tutorial
    2. How to Implement PushAuth™: iOS Mobile App tutorial
    3. How to Implement PushAuth™: Android Mobile App tutorial
  3. Building a Web Application with PushAuth™: Technical deep-dive into the details of building a web application with PushAuth™ integrated.
  4. The Power of Trusted Registration: A more secure extension of the simple user sign-up flow that includes user registration and trusted device registration.
  5. Integrating Trusted Registration for PushAuth™: Technical deep-dive into the details of extending the project with trusted device registration.

Push authentication is one of the most seamless and secure methods of multi-factor authentication. We want to help you understand its importance and the power it has to upgrade your security!

Table Stakes

Let’s clear up two terms that are critical to understanding what PushAuth is solving:

Authorization = allowing you access to a system

Authentication = verifying your identity

There is a great Medium article that describes those differences in more depth. You could have an authorization flow that requires no authentication, letting all users in without proving anything (please don’t do this). On the other hand, you could have an authorization flow that makes users jump through countless hoops to try to make 100% sure the true user is accessing their account. The magic is in striking a balance between these extremes, ensuring a secure authorization process without excessive user annoyance.

One more term to define before moving forward:

Multi-factor authentication = an authentication method that requires two or more pieces of proof before granting a user access to resources

Multi-factor authentication (MFA) is sometimes referred to as two-factor authentication (2FA). Here is a page on MFA basics provided by NIST.

Why use multi-factor authentication?

Before jumping into push authentication, let’s explain MFA. The three basic authentication mechanisms are:

  1. Knowledge: something you know (password, security question answers, PIN)
  2. Possession: something you have (access badge, ATM card, smartphone, hard token)
  3. Biometric: something you are (fingerprint, face, gait, voice)

As defined above, MFA uses two or more of these authentication mechanisms. There are many security benefits to using MFA. The primary benefit is increased confidence in an authorization flow. This stronger assertion on the user’s identity increases the application’s security. Using MFA mitigates the dangers of poor security habits many users have with passwords, such as writing them down, making them simple or easily-guessed, and reusing them across platforms.

Push authentication is one method of accomplishing MFA. This method usually requires a user to download an app on their phone and register it with their account. The mobile app receives push notifications during the login flow. In doing so, the user establishes two authentication mechanisms: knowledge of their username/password combination and possession of their registered device. Here is an example of using push authentication in a login flow:

A user logs in to a web application by entering their username and password. Upon clicking the button to log in, the server sends a push notification to the app (which the user has already downloaded and registered). The user unlocks their phone and authorizes the request by clicking “Accept” on the notification. The server receives this approved response and proceeds to log the user in to the web app. If the user instead clicks “Deny,” the authentication step fails and the user would not gain access to the web app.

Other types of MFA include SMS codes, phone calls, email codes or magic links, authenticator app verification codes, and hardware tokens. This PCMag article does a good job of aggregating different ways well-known companies offer MFA. In the next section we discuss the advantages push authentication has over other methods.

Why choose push authentication?

In a nutshell, push authentication increases security in a cost-effective manner without affecting usability. This section lists some of the aspects specific to push authentication that make it superior to other methods of MFA.

MFA methodFree tierCost per million
SMS100$6,540
Email1,000$20
Mobile Push Notifications1 million$0.50
  • Convenient and easy to use. We can assume that users who have smartphones are comfortable with using apps and interacting with notifications. This makes using push authentication easy for them, since there is nothing else they need to learn to use or get used to the method. Push notifications are much easier to respond to – there is no typing, clicking links, or copying codes while trying to beat the timer. Authentication happens with a simple and speedy tap on their screen.
  • Secure. The push authentication mobile app is installed securely on users’ devices, so there is no reliability on a user’s accounts with external companies. This out-of-band communication can’t be intercepted at the point of password entry, and it is encrypted from end to end between the application and the secured push authentication provider. The many insecurities of the SMS method led to the downfall of its prominence in this space.
  • Fast. Since authentication requests are sent in real time via notifications, a user can become aware of and deny fraudulent requests when they happen and promptly take action. Unlike some methods of MFA, the user must unlock their device before responding to the request. However, this is a level of friction that users are already familiar with on a day-to-day (or, more realistically, minute-to-minute) basis. For devices with TouchID or FaceID, this process is almost frictionless. This even increases the security by preventing unhindered access to authentication by an attacker.
  • No hardware to manage. Unlike with hardware tokens, push authentication utilizes users’ existing phones. Since most people consider their phone an extension of their body, this also means the hardware is unlikely to be easily or frequently misplaced. When phones are lost or broken, the responsibility is on the user, rather than the service provider, to obtain a replacement.
  • No over-the-shoulder copy-ability. Codes and magic links sent via SMS/email and codes generated in an authenticator app can all be intercepted and used to log in as an attacker. Push notifications don’t contain anything that can be copied or reproduced.

Is push authentication perfect?

Unsurprisingly, like most things, no. On a base level, push authentication requires that every user has a fully-functional smartphone. Without a smartphone, they can’t have the mobile app (and the provider needs to have an application). Without internet connection, their device can’t receive push notifications through the mobile app. In the specific cases of a user’s phone being lost, stolen, or otherwise not in their possession, push authentication will fail. Even if the device is in their possession but out of battery, waterlogged, or otherwise not functional, the same assertion applies. For the sake of argument, let’s assume that none of these scenarios are issues. Push authentication is still not “perfect.”

There are some valid security concerns surrounding the use of push authentication as a method for MFA. Smartphones themselves, where the authentication bit happens, are vulnerable to attacks and viruses. There’s also the issue of trusting a device to be associated with its true user during registration, as well as not allowing attackers to register devices under the true user’s account. Even with a trusted device, any pre-registered device is a weak point for attacker entry. You can’t ensure that users have their devices protected via a locking mechanism. If they don’t have access into their phone locked, then it’s pretty easy for an attacker to accept a notification on the user’s behalf. Finally, it’s completely possible that a user accidentally approves a fraudulent request on their device without realizing it, or realizing too late.

However, there are solutions to these problems. In the cases of a temporarily out-of-possession device, you can provide fallback resources (one-time passcodes, email or SMS delivery, security questions, etc.) as a workaround for that authentication attempt. If the device is permanently out of possession, such as a ruined or stolen phone, you can provide a method of revoking access from the old device and allowing users to self-sign up and provision the app on a new mobile device. If an attacker gains physical access to the device before access is revoked, it’s likely that the user will have yet another authentication barrier in the form of a passcode, TouchID, or FaceID. While this can be attacked as well, it still raises the bar for entry before access is revoked. With regards to users accepting fraudulent requests on their device, the push notification prompt could be extended to include a unique identifier that is expected to match a value on the web login page. The existence of this would set the expectation that a user is to confirm the value before accepting the request.

However, there’s only so much you can do as a developer. People will still make mistakes, and hackers will still gain access. The main point here is that these are things you, as a developer, should be aware of and address. All things considered – push authentication sits in a pretty great place security-wise, is incredibly low friction, and (perhaps most importantly) incredibly cost-effective. Keep an eye out for our next post of the series, which will walk you through using our open-source project to implement PushAuth yourself!