Gemini Enterprise
Discover Gemini Enterprise engines and assistants in a GCP project.
Airia discovers Gemini Enterprise deployments in a GCP project: it enumerates the Gemini Enterprise engines in the project and the assistants defined in each engine.
This is its own connector, distinct from the Google Workspace connector (domain-level visibility through the Workspace Admin SDK) and from the Google Agent Platform connector (Vertex AI Agent Engine and Dialogflow CX).
Authentication model
A GCP service account JSON key. Airia authenticates with the
https://www.googleapis.com/auth/cloud-platform scope and reads through the Discovery
Engine API (the API behind Gemini Enterprise), listing engines and their assistants.
Required GCP-side configuration
| Requirement | Value |
|---|---|
| API enabled on the project | Discovery Engine API |
| IAM role for the service account | Discovery Engine Viewer (roles/discoveryengine.viewer), the least-privilege read role for engines and assistants |
Customer-side setup steps
- In the GCP Console, open the project hosting your Gemini Enterprise deployment.
- Confirm the Discovery Engine API is enabled (APIs & Services → Library).
- IAM & Admin → Service Accounts → Create service account (for example
airia-gemini-discovery). - Grant it Discovery Engine Viewer (
roles/discoveryengine.viewer) on the project. - Create a JSON key for the service account and download it.
- In Airia, go to Discover → Connections → Gemini Enterprise and create a new connection.
Form fields the Airia UI asks for
| Field | Required | Notes |
|---|---|---|
| Project ID | Yes | The GCP project ID |
| Location | Yes | global for most Gemini Enterprise deployments, or the region it lives in |
| Service account JSON | Yes | The full JSON key, pasted as-is. Stored encrypted, masked on edit |
Multiple environments
Multiple Gemini Enterprise connections per tenant are supported: one per GCP project and location pair.