canonical: https://jentic.com/apis/googleapis.com/iap

# Google Cloud Identity-Aware Proxy API

The Cloud Identity-Aware Proxy (IAP) API manages the configuration that fronts internal applications on Google Cloud with identity- and context-based access control. It is used to register OAuth brands and clients, configure per-resource IAP settings (such as access expressions and IAP-TCP tunnel options), maintain destination groups for TCP forwarding, validate attribute expressions, and reset OAuth client secrets. IAP itself enforces access at the load balancer in front of HTTP backends and Compute Engine VMs; this API is the control plane that defines how that enforcement behaves.

## For AI agents

Configure Identity-Aware Proxy for Google Cloud apps and VMs: OAuth brands and clients, per-resource IAP settings, destination groups, and attribute expressions.

## Scope

Does not authenticate end users at runtime, route HTTP traffic, or run TCP tunnels - use for IAP control-plane configuration of brands, clients, settings, and destination groups only.

## Capabilities

- Create and manage OAuth brands that back IAP-protected applications
- Register, list, and reset OAuth client secrets used by IAP for sign-in
- Read and update per-resource IAP settings, including access expressions and TCP tunnel options
- Manage destination groups that scope IAP TCP forwarding to specific internal hosts and ports
- Validate IAP attribute expressions before deploying access rules
- Delete IAP resources and brands during teardown of protected applications

## Use cases

### Front Internal Apps with Identity-Based Access

Platform teams put IAP in front of internal admin tools so only signed-in users from approved groups can reach them. They use the IAP API to update each backend service's iapSettings (access expressions referencing groups, IP ranges, or device posture) and reset the OAuth client secret on rotation. The setup replaces VPN-only access for many internal tools.

Example prompt: For backend service projects/p/iap_web/compute/services/internal-admin, call PATCH /v1/{+name}:iapSettings to set accessSettings.allowedDomains to a list containing example.com and return the updated iapSettings.

### IAP TCP Forwarding to Bastion-less SSH

Operations teams configure IAP TCP forwarding to give engineers SSH-over-IAP access to internal VMs without running a bastion host. Destination groups created via the API restrict which hosts and ports a given IAP-protected resource can reach, so policies are explicit and auditable.

Example prompt: Create a destination group named ssh-prod with destinations for hosts of pattern 10.0.1.0/24 and ports 22 under projects/p, and confirm the group appears in GET /v1/{+parent}/destGroups.

### OAuth Brand and Client Lifecycle

When standing up a new IAP-protected app, teams use the API to register an OAuth brand for the project, create an identity-aware proxy client, and rotate its secret on schedule. Brands and clients are listed by parent for inventory, and resetSecret rotates the client secret used by IAP for sign-in to that app.

Example prompt: Under projects/p, POST a new brand with applicationTitle Internal-Admin, then POST an identity-aware proxy client under that brand named admin-web and return the resulting client_id and secret.

### AI Agent IAP Settings Audit

An AI agent reviewing access controls calls the IAP API via Jentic to enumerate every IAP-protected resource in a project and validate that access expressions match the declared policy. Drift is reported with a suggested PATCH payload. Jentic isolates the IAP admin credential so the agent never holds raw service-account JSON.

Example prompt: List IAP settings across resources under projects/p by calling GET /v1/{+name}:iapSettings for each service, and emit a CSV of resource, accessSettings.gcipSettings, and accessSettings.allowedDomains.

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | /v1/{+name}:iapSettings | Get IAP settings for a resource |
| PATCH | /v1/{+name}:iapSettings | Update IAP settings for a resource |
| POST | /v1/{+name}:resetSecret | Reset the OAuth client secret for an identity-aware proxy client |
| POST | /v1/{+name}:validateAttributeExpression | Validate an IAP attribute expression |
| GET | /v1/{+parent}/brands | List OAuth brands under a project |
| POST | /v1/{+parent}/brands | Create an OAuth brand |
| GET | /v1/{+parent}/destGroups | List destination groups for IAP TCP forwarding |
| POST | /v1/{+parent}/destGroups | Create a destination group for IAP TCP forwarding |

## Key resources

- **Brands** — Register and list OAuth brands that back IAP-protected applications
- **IdentityAwareProxyClients** — Create, list, and reset OAuth clients for IAP sign-in
- **IapSettings** — Read and update per-resource IAP behaviour (access expressions, TCP tunnel options)
- **DestGroups** — Manage destination groups for IAP TCP forwarding

## Why Jentic

- **Setup:** Wiring the Cloud Identity-Aware Proxy API by hand means configuring admin OAuth2, narrowing to the cloud-platform scope, refreshing tokens, and URL-encoding its brand, client, and settings resource paths yourself. Through Jentic you install once, import the Cloud Identity-Aware Proxy API from the API Directory, store the credential once, and your agent calls it.
- **Permission scoping:** Resource names travel in the URL path (/v1/{name}:iapSettings, /v1/{parent}/brands), so a rule can pin your agent to one project's IAP configuration: it reads and updates that project's settings and nothing else. You choose the operations it may call, so a client-secret reset is not included unless you add it.
- **Credential handling:** Your IAP admin credential is stored once, encrypted, by your own Jentic One instance and injected at execution time as a short-lived cloud-platform access token. The service-account key never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'update iap settings' or 'list oauth brands', and Jentic returns the matching operation with its input schema so the agent calls the right endpoint without browsing the URL-encoded resource-name conventions.

## Related APIs

- **Identity and Access Management (IAM) API** — IAM grants the iap.* roles that authorise calls to the IAP API and the IAP-secured-web-app-user role at the resource.
- **IAM Service Account Credentials API** — Mints OIDC ID tokens used to authenticate service-to-service calls behind IAP.
- **Access Context Manager API** — Defines context-based access levels that IAP access expressions can reference.

## FAQ

### What authentication does the Cloud IAP API use?

Google OAuth 2.0 with the cloud-platform scope and an IAM principal that holds iap.* permissions on the resource being managed. Through Jentic the credential is stored encrypted and exchanged for a scoped access token at execution.

### Can I update access policy on an IAP-protected service?

Yes. Call PATCH /v1/{+name}:iapSettings on the resource (a backend service, Cloud Run service, or App Engine app) with the desired accessSettings, including allowedDomains, gcipSettings, and corsSettings. The new settings take effect within minutes.

### What are the rate limits for the IAP API?

Google enforces per-project quotas on IAP admin operations (default in the low hundreds of QPS). Settings updates are eventually consistent; callers should expect propagation latency of seconds to a few minutes.

### How do I rotate an IAP client secret through Jentic?

Search Jentic with 'reset iap client secret', load the POST /v1/{+name}:resetSecret operation, and execute with name set to projects/{p}/brands/{brand}/identityAwareProxyClients/{clientId}. Jentic returns the new client secret, which can then be wired into your sign-in configuration.

### Does this API issue tokens for end users hitting an IAP-protected app?

No. End-user sign-in to an IAP-protected app is handled by IAP itself at the load balancer. This API is the control plane for IAP configuration (brands, clients, settings, destination groups), not the runtime token issuer.

### Can I limit what my agent is allowed to do with the Cloud Identity-Aware Proxy API?

Yes. Because you run Jentic One yourself, your own rules decide which IAP operations the agent may call and which credential it uses. Since resource names travel in the URL path, such as /v1/{name}:iapSettings and /v1/{parent}/brands, you can pin the agent to one project's IAP configuration so it only reads and updates that project's settings, brands, and destination groups. You choose the operations it can reach, so a client-secret reset via :resetSecret stays off unless you explicitly add it.