Add Kubeflow OAuth2 application

app-kubeflow.tf

@@ -0,0 +1,50 @@
+# =============================================================================
+# Kubeflow - ML Platform Dashboard
+# =============================================================================
+
+data "authentik_property_mapping_provider_scope" "kubeflow" {
+  managed_list = [
+    "goauthentik.io/providers/oauth2/scope-openid",
+    "goauthentik.io/providers/oauth2/scope-email",
+    "goauthentik.io/providers/oauth2/scope-profile",
+  ]
+}
+
+resource "authentik_provider_oauth2" "kubeflow" {
+  name               = "Kubeflow"
+  client_id          = "kubeflow"
+  client_type        = "confidential"
+  authorization_flow = data.authentik_flow.default_authorization.id
+  invalidation_flow  = data.authentik_flow.default_invalidation.id
+  
+  access_token_validity  = "hours=1"
+  refresh_token_validity = "days=30"
+  
+  property_mappings = data.authentik_property_mapping_provider_scope.kubeflow.ids
+  
+  allowed_redirect_uris = [
+    { matching_mode = "strict", url = "https://kubeflow.walleye-frog.ts.net/oauth2/callback" },
+  ]
+  
+  signing_key = data.authentik_certificate_key_pair.generated.id
+}
+
+resource "authentik_application" "kubeflow" {
+  name              = "Kubeflow"
+  slug              = "kubeflow"
+  protocol_provider = authentik_provider_oauth2.kubeflow.id
+  
+  meta_description = "ML Training Platform"
+  meta_launch_url  = "https://kubeflow.walleye-frog.ts.net"
+  
+  group = "DevOps"
+}
+
+output "kubeflow_client_id" {
+  value = authentik_provider_oauth2.kubeflow.client_id
+}
+
+output "kubeflow_client_secret" {
+  value     = authentik_provider_oauth2.kubeflow.client_secret
+  sensitive = true
+}