Protect your Firebase ML Apple app's Cloud credentials

If your Apple app uses one of Firebase ML's cloud APIs, before you launch your app in production, you should take some additional steps to prevent unauthorized API access.

1. Reduce the scope of existing API keys

First, configure your existing API keys to disallow access to the Cloud Vision API:

  1. Open the Credentials page of the Google Cloud console. When prompted, select your project.

  2. For each existing API key in the list, open the editing view.

  3. In the API restrictions section, select Restrict key, then add to the list all of the APIs to which you want the API key to have access. Make sure to not include the Cloud Vision API.

    When you configure an API key's API restrictions, you are explicitly declaring the APIs to which the key has access. By default, when the API restrictions section has Don't restrict key selected, an API key can be used to access any API that is enabled for the project.

Now, your existing API keys will not grant access to cloud ML services, but each key will continue to work for any APIs that you added to its API restrictions list.

Note that if you enable any additional APIs in the future, you must add them to the API restrictions list for the applicable API key.

2. Create a new API key for use with Firebase ML

Next, create a new API key for Firebase ML that only allows calls to the Cloud Vision API:

  1. Return to the Credentials page. Be sure your Firebase project is still selected.

  2. Click Create credentials > API key. Take note of the new API key, then click Restrict key.

  3. In the API restrictions section, select Restrict key, then add to the list only the Cloud Vision API.

This API key grants access only to the Cloud Vision API and can be used by Firebase ML to access cloud-based models.

To help mitigate the consequences of a compromised key, you should reduce the Cloud Vision API's per-user quota from its default setting. To do so:

  1. Open the Cloud Vision API Quotas page of the Google Cloud console. When prompted, select your project.

  2. In the Requests section, set the Requests per minute per user quota to something reasonable for your app. For example, if your app involves uploading a picture of a document to get back its text, it's unlikely that a user will do that more than once every few seconds, so a quota of 30-40 would probably be safe.

    Note that in this context "requests per user" refers to requests from a single IP address. You might need to consider this if you expect multiple users to use your app at the same time from behind a NAT.

4. Call Cloud APIs using your Firebase ML API key

Finally, in your app, configure Firebase ML to use your new API key.

Because the Firebase ML API key allows unauthenticated access to the Cloud Vision API, it's important to keep the key confidential to prevent unauthorized use and charges to your billing account. To do so, you should refrain from including your API key in your app binary. Instead, at app runtime, verify that a known good user is signed in, and only then, retrieve the API key from a server.

Even when these practices are observed, it is possible for an API key to be compromised. You should take steps to help mitigate the consequences of a compromised key, such as reducing the API's per-user quota as described above, implementing key rotation policies, and issuing different keys to different groups of users.

After your app has safely acquired the API key, when you want to call an Firebase ML Cloud API, specify the key:


if let cloudVisionKey = getYourApiKey() {  // See note above about securing your API key
    let options = VisionCloudDetectorOptions()
    options.apiKeyOverride = cloudVisionKey
    let cloudDetector = options)


NSString *cloudVisionKey = [self getYourApiKey];  // See note above about securing your API key
if (cloudVisionKey != nil) {
    FIRVisionCloudDetectorOptions *options =
            [[FIRVisionCloudDetectorOptions alloc] init];
    options.APIKeyOverride = cloudVisionKey;
    FIRVisionCloudLandmarkDetector *landmarkDetector =
            [vision cloudLandmarkDetectorWithOptions:options];

In addition, you should follow the general advice in Securing an API key.

Next steps

See the launch checklist for information on preparing your app to launch when using other Firebase features.