Unable to install @azure/ai-vision-face-ui — missing dependency @azure-ai-vision-face/ui-assets

Hi Infra | Ilitera

The npm installation for @azure/ai-vision-face-ui fails because it depends on a private companion package @azure-ai-vision-face/ui-assets@1.4.5. This assets package is not published on the public npm registry and appears missing or unpublished in the Azure DevOps private feed. As a result, npm install returns a 404 Not Found error for the dependency.

Nature of the Package

Unlike standard Azure SDKs, the Face UI Web SDK uses a gated distribution model. The assets package is hosted in a private Azure DevOps feed and requires authentication using a Client Assets Access Token (not a regular Azure AD token).

The official README instructs developers to configure .npmrc with this token before installing the UI package.

There are two practical options to unblock development:

• Option A: Pin the UI package to an older version (e.g., 1.4.2) that matches an available assets version in the feed. After installation, manually copy the assets folder into your app’s public/ directory.
• Option B: Skip npm resolution for the missing assets and host them manually. Use the Client Assets Access Token API to download the assets bundle and place it under public/facelivenessdetector-assets/. This approach works even if the ui-assets@1.4.5 package is not resolvable.

Configuring Authentication

To access the private feed, create or update your .npmrc file with:

registry=https://pkgs.dev.azure.com/msface/SDK-packaging/AzureAIVision/npm/registry/ always-auth=true //pkgs.dev.azure.com/msface/SDK-packaging/AzureAIVision/npm/registry/:username=msface //pkgs.dev.azure.com/msface/SDK-packaging/AzureAIVision/npm/registry/:_password=<BASE64_ACCESS_TOKEN> //pkgs.dev.azure.com/msface/SDK-packaging/AzureAIVision/npm/registry/:email=<placeholder>

The <BASE64_ACCESS_TOKEN> comes from the Get Client Assets Access Token endpoint, which requires Face Liveness gating approval.

Checklist for Resolution

1. Confirm your subscription is approved for Face Liveness (it’s GA but gated).
2. Obtain a fresh Client Assets Access Token.
3. Configure .npmrc for the private feed.
4. Check available versions with npm view @azure-ai-vision-face/ui-assets --registry <feed-url>.
5. If 1.4.5 is missing, use Option A or Option B above.
6. For UI functionality, host assets manually if needed.
7. For backend-only use, switch to @azure-rest/ai-vision-face

Reference:

https://learn.microsoft.com/en-us/azure/ai-services/cognitive-services-container-support

https://learn.microsoft.com/en-us/python/api/overview/azure/cognitiveservices/face-readme?view=azure-python

https://learn.microsoft.com/en-us/azure/machine-learning/how-to-configure-cli?view=azureml-api-2&tabs=public

https://learn.microsoft.com/en-us/azure/ai-foundry/foundry-models/concepts/models-sold-directly-by-azure?view=foundry-classic&tabs=global-standard-aoai%2Cstandard-chat-completions%2Cglobal-standard&pivots=azure-openai

Thank you!

Hi Infra | Ilitera

Did you get any chance to review the above response. Thank you!