Google Agent Platform
Discover Vertex AI Agent Engine and Dialogflow CX agents with one GCP connection.
Airia's Google Agent Platform connector discovers agents on two GCP product surfaces with a single connection:
- Dialogflow CX agents
- Vertex AI Agent Engine (reasoning engines / agent workflows)
GCP coverage at a glance
Three questions that often get conflated; keep them separate:
| Question | Answer |
|---|---|
| Can Airia discover Vertex AI Agent Engine and Dialogflow CX agents? | Yes, via this connector |
| Can Airia route Gemini models through the AI Gateway? | Yes, via the Gemini gateway provider (public Generative AI API, API-key auth). Configured separately from this connector |
| Can Airia route Vertex-resident models (Gemini on Vertex, Claude on Vertex) through the AI Gateway with GCP service-account auth? | Not currently supported |
Authentication model
A GCP service account JSON key for the project hosting your Dialogflow and Vertex resources.
Required GCP IAM roles (least privilege)
Grant the service account both roles on the project:
| Role | Covers |
|---|---|
roles/dialogflow.reader (Dialogflow API Reader) | Listing and reading Dialogflow CX agents (dialogflow.agents.list, dialogflow.agents.get) |
roles/aiplatform.viewer (Vertex AI Viewer) | Listing and reading Vertex Agent Engine reasoning engines (aiplatform.reasoningEngines.list, aiplatform.reasoningEngines.get) |
These two roles together cover everything the connector calls. If your internal policy
prefers a single broader role, roles/aiplatform.user (Vertex AI User) also works but grants
more than needed.
Customer-side setup steps
- In the GCP Console, open the project that hosts your Dialogflow CX agents or Vertex AI Agent Engine deployments.
- IAM & Admin → Service Accounts → Create service account. Name it something like
airia-agent-discovery. - Grant the service account the two roles above on the project.
- Create a JSON key for the service account and download it.
- In Airia, go to Discover → Connections → Google Agent Platform and create a new connection with the project ID, location, and the pasted JSON key.
Form fields the Airia UI asks for
| Field | Required | Notes |
|---|---|---|
| Project ID | Yes | The GCP project hosting your Dialogflow / Vertex resources |
| Location | Yes | The region the resources live in (for example us-central1) |
| Service account JSON | Yes | The full JSON key, pasted as-is. Stored encrypted, masked on edit |
Multiple environments
Multiple Google Agent Platform connections per tenant are supported: typically one per GCP project and location pair.
Related connectors
- Gemini Enterprise is its own connector with the same field shape (project ID, location, service account JSON); see the Gemini Enterprise connector.
- Google Workspace covers domain-level AI deployments and per-user OAuth grants; see Google Workspace. It is a different connector with different credentials.
Was this page helpful?