Gemini Enterprise publishes custom UI authentication guide
Understand this story
Google published a guide explaining that when enterprises build their own Customer UI to call Gemini Enterprise's StreamAssist API, they must take over the authentication flow that the built-in web app originally handled automatically, and it provides the architecture, component flows and runnable code for three layers of authentication: user, connector and agent. The guide lists the conditions for use: all principals calling the StreamAssist API must be granted the Discovery Engine User IAM role on the GCP project; when using the human-user-facing patterns (Pattern 1A, 1B), the user or Workforce Pool group must also have a valid Gemini Enterprise user license in the Admin Console. If the Customer UI is deployed on a custom domain, you need to work with the Gemini Enterprise team to enable the custom origin flag so that the callback address is returned automatically and the popup closes.
AI overview based on source reports · Updated 56 min ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Google Developers BlogHow to handle user, connector, and agent auth in Gemini Enterprise with a custom UI
This guide explains how enterprises building a custom UI that calls the Gemini Enterprise StreamAssist API must handle three authentication layers themselves: user or workload identity to StreamAssist (Google Identity, Workforce Identity Federation, or Workload Identity Federation), Federated Connector 3-legged OAuth via the managed Keystore, and on-demand per-tool OAuth for custom ADK agents on Vertex AI Agent Engine using GcpAuthProvider and the GCP Auth Manager Token Vault. It details the component flows, IAM roles and licensing requirements, and a runnable code blueprint in the gemini-enterprise-custom-ui-auth GitHub repository.
Activity for this story
Not enough continuous observations to show a trend yet.