Feat: add observedGeneration; simplify conditions

charts/kagenti-operator/crds/agent.kagenti.dev_agentruntimes.yaml

@@ -26,14 +26,9 @@
       jsonPath: .spec.targetRef.name
       name: Target
       type: string
-    - description: Runtime Phase
-      jsonPath: .status.phase
-      name: Phase
-      type: string
-    - description: Card Fetch Status
-      jsonPath: .status.conditions[?(@.type=='CardFetched')].status
-      name: CardFetched
-      priority: 1
+    - description: Ready Status
+      jsonPath: .status.conditions[?(@.type=='Ready')].status
+      name: Ready
       type: string
     - jsonPath: .metadata.creationTimestamp
       name: Age
@@ -425,7 +420,7 @@
                     lastTransitionTime:
                       description: |-
                         lastTransitionTime is the last time the condition transitioned from one status to another.
                         This should be when the underlying condition changed.  If that is not known, then using the time when the API field changed is acceptable.
                       format: date-time
                       type: string
                     message:
@@ -437,7 +432,7 @@
                     observedGeneration:
                       description: |-
                         observedGeneration represents the .metadata.generation that the condition was set based upon.
                         For instance, if .metadata.generation is currently 12, but the .status.conditions[x].observedGeneration is 9, the condition is out of date
                         with respect to the current state of the instance.
                       format: int64
                       minimum: 0
@@ -486,13 +481,11 @@
                 items:
                   type: string
                 type: array
-              phase:
-                description: Phase is the high-level state of the AgentRuntime
-                enum:
-                - Pending
-                - Active
-                - Error
-                type: string
+              observedGeneration:
+                description: ObservedGeneration is the most recent generation observed
+                  by the controller.
+                format: int64
+                type: integer
             type: object
         required:
         - spec

kagenti-operator/GETTING_STARTED.md

@@ -141,8 +141,8 @@ The controller will:
 kubectl get agentruntime -n team1
 
 # Example output:
-# NAME                      TYPE    TARGET          PHASE    AGE
-# weather-agent-runtime     agent   weather-agent   Active   2m
+# NAME                      TYPE    TARGET          READY   AGE
+# weather-agent-runtime     agent   weather-agent   True    2m
 
 # View detailed conditions
 kubectl describe agentruntime weather-agent-runtime -n team1

kagenti-operator/api/v1alpha1/agentruntime_types.go

@@ -28,15 +28,6 @@ const (
 	RuntimeTypeTool  RuntimeType = "tool"
 )
 
-// +kubebuilder:validation:Enum=Pending;Active;Error
-type RuntimePhase string
-
-const (
-	RuntimePhasePending RuntimePhase = "Pending"
-	RuntimePhaseActive  RuntimePhase = "Active"
-	RuntimePhaseError   RuntimePhase = "Error"
-)
-
 // +kubebuilder:validation:Enum=mtls;http
 type TransportSecurity string
 
@@ -188,9 +179,9 @@ type CardStatus struct {
 
 // AgentRuntimeStatus defines the observed state of AgentRuntime.
 type AgentRuntimeStatus struct {
-	// Phase is the high-level state of the AgentRuntime
+	// ObservedGeneration is the most recent generation observed by the controller.
 	// +optional
-	Phase RuntimePhase `json:"phase,omitempty"`
+	ObservedGeneration int64 `json:"observedGeneration,omitempty"`
 
 	// ConfiguredPods is the count of pods with expected labels/config
 	// +optional
@@ -217,8 +208,7 @@ type AgentRuntimeStatus struct {
 // +kubebuilder:resource:shortName=art;agentrt
 // +kubebuilder:printcolumn:name="Type",type="string",JSONPath=".spec.type",description="Workload Type"
 // +kubebuilder:printcolumn:name="Target",type="string",JSONPath=".spec.targetRef.name",description="Target Workload"
-// +kubebuilder:printcolumn:name="Phase",type="string",JSONPath=".status.phase",description="Runtime Phase"
-// +kubebuilder:printcolumn:name="CardFetched",type="string",JSONPath=".status.conditions[?(@.type=='CardFetched')].status",description="Card Fetch Status",priority=1
+// +kubebuilder:printcolumn:name="Ready",type="string",JSONPath=".status.conditions[?(@.type=='Ready')].status",description="Ready Status"
 // +kubebuilder:printcolumn:name="Age",type="date",JSONPath=".metadata.creationTimestamp"
 
 // AgentRuntime attaches runtime configuration to a backing workload classified as an

kagenti-operator/config/crd/bases/agent.kagenti.dev_agentruntimes.yaml

@@ -26,14 +26,9 @@ spec:
       jsonPath: .spec.targetRef.name
       name: Target
       type: string
-    - description: Runtime Phase
-      jsonPath: .status.phase
-      name: Phase
-      type: string
-    - description: Card Fetch Status
-      jsonPath: .status.conditions[?(@.type=='CardFetched')].status
-      name: CardFetched
-      priority: 1
+    - description: Ready Status
+      jsonPath: .status.conditions[?(@.type=='Ready')].status
+      name: Ready
       type: string
     - jsonPath: .metadata.creationTimestamp
       name: Age
@@ -486,13 +481,11 @@ spec:
                 items:
                   type: string
                 type: array
-              phase:
-                description: Phase is the high-level state of the AgentRuntime
-                enum:
-                - Pending
-                - Active
-                - Error
-                type: string
+              observedGeneration:
+                description: ObservedGeneration is the most recent generation observed
+                  by the controller.
+                format: int64
+                type: integer
             type: object
         required:
         - spec

kagenti-operator/docs/api-reference.md

@@ -462,7 +462,7 @@ The AgentRuntime controller applies the following labels and annotations to the
 
 | Field | Type | Description |
 |-------|------|-------------|
-| `phase` | string | High-level state of the AgentRuntime (`Pending`, `Active`, or `Error`) |
+| `observedGeneration` | int64 | Most recent generation observed by the controller |
 | `configuredPods` | int32 | Count of pods with expected labels/configuration |
 | `conditions` | [][Condition](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.28/#condition-v1-meta) | Current state of the AgentRuntime |
 
@@ -480,7 +480,6 @@ The AgentRuntime controller applies the following labels and annotations to the
 | `IstioMeshEnrolled` | True | `NamespaceLabeled` | Namespace labeled with `istio-discovery=enabled` and `istio.io/dataplane-mode=ambient` for Istio ambient mesh enrollment |
 | `IstioMeshEnrolled` | False | `OptedOut` | Namespace has `kagenti.io/istio-mesh=disabled` annotation; Istio mesh labels not applied |
 | `IstioMeshEnrolled` | False | `PatchFailed` | Failed to patch namespace labels (e.g., RBAC misconfiguration). Non-fatal; reconcile continues. |
-| `SkillsDiscovered` | True | `SkillsFound` | Linked skills discovered from `kagenti.io/skills` annotation on the target workload |
 | `SkillsMounted` | True | `SkillsApplied` | OCI skill ImageVolumes applied to the target workload |
 | `SkillsMounted` | False | `FeatureGateDisabled` | Skills defined but `skillImageVolumes` feature gate is disabled |
 | `SkillsMounted` | False | `UnsupportedWorkloadKind` | Skills defined but the target workload kind (e.g., Sandbox) does not support skill ImageVolumes |
@@ -586,16 +585,13 @@ kubectl get art
 kubectl get agentruntimes
 
 # Example output:
-# NAME                      TYPE    TARGET          PHASE    AGE
-# weather-agent-runtime     agent   weather-agent   Active   5m
-# calculator-tool-runtime   tool    calculator-tool Active   3m
+# NAME                      TYPE    TARGET          READY   AGE
+# weather-agent-runtime     agent   weather-agent   True    5m
+# calculator-tool-runtime   tool    calculator-tool True    3m
 
 # Get detailed information
 kubectl describe agentruntime weather-agent-runtime
 
-# View runtime phase
-kubectl get art weather-agent-runtime -o jsonpath='{.status.phase}'
-
 # View configured pods count
 kubectl get art weather-agent-runtime -o jsonpath='{.status.configuredPods}'
 ```

kagenti-operator/internal/controller/agentruntime_config_test.go

@@ -334,9 +334,6 @@ var _ = Describe("AgentRuntime Config", func() {
 			updated := &agentv1alpha1.AgentRuntime{}
 			Expect(k8sClient.Get(ctx, nn, updated)).To(Succeed())
 
-			// Should still be Active (warning doesn't block reconciliation)
-			Expect(updated.Status.Phase).To(Equal(agentv1alpha1.RuntimePhaseActive))
-
 			// Should have ConfigResolved condition with warning
 			var configCond *metav1.Condition
 			for i := range updated.Status.Conditions {

kagenti-operator/internal/controller/agentruntime_controller.go

@@ -69,11 +69,9 @@ const (
 	AnnotationRestartPending = "kagenti.io/restart-pending"
 
 	// Condition types for AgentRuntime status.
-	ConditionTypeReady            = "Ready"
-	ConditionTypeTargetResolved   = "TargetResolved"
-	ConditionTypeConfigResolved   = "ConfigResolved"
-	ConditionTypeCardFetched      = "CardFetched"
-	ConditionTypeSkillsDiscovered = "SkillsDiscovered"
+	ConditionTypeReady          = "Ready"
+	ConditionTypeTargetResolved = "TargetResolved"
+	ConditionTypeConfigResolved = "ConfigResolved"
 
 	// AnnotationLastCardFetchHash stores the change-detection key used to skip
 	// redundant card fetches when the workload's pod template has not changed.
@@ -137,6 +135,8 @@ func (r *AgentRuntimeReconciler) Reconcile(ctx context.Context, req ctrl.Request
 		return ctrl.Result{}, client.IgnoreNotFound(err)
 	}
 
+	rt.Status.ObservedGeneration = rt.Generation
+
 	// 2. Handle deletion
 	if !rt.DeletionTimestamp.IsZero() {
 		return r.handleDeletion(ctx, rt)
@@ -213,11 +213,7 @@ func (r *AgentRuntimeReconciler) Reconcile(ctx context.Context, req ctrl.Request
 		if r.Recorder != nil {
 			r.Recorder.Event(rt, corev1.EventTypeWarning, "SCCBindingError", err.Error())
 		}
-		r.setPhase(rt, agentv1alpha1.RuntimePhaseError)
-		r.setCondition(rt, ConditionTypeReady, metav1.ConditionFalse, "SCCBindingError", err.Error())
-		if statusErr := r.Status().Update(ctx, rt); statusErr != nil {
-			logger.Error(statusErr, "Failed to update status")
-		}
+		r.updateErrorStatus(ctx, req.NamespacedName, ConditionTypeReady, "SCCBindingError", err.Error())
 		return ctrl.Result{RequeueAfter: 30 * time.Second}, nil
 	}
 
@@ -268,20 +264,12 @@ func (r *AgentRuntimeReconciler) Reconcile(ctx context.Context, req ctrl.Request
 
 	// 8. Update status (retry on conflict to preserve all conditions computed above)
 	rt.Status.ConfiguredPods = configuredPods
-	r.setPhase(rt, agentv1alpha1.RuntimePhaseActive)
 	r.setCondition(rt, ConditionTypeReady, metav1.ConditionTrue, "Configured",
 		fmt.Sprintf("Workload %s configured with config-hash %s", rt.Spec.TargetRef.Name, configResult.Hash[:12]))
 	if fg.SkillDiscovery {
 		rt.Status.LinkedSkills = linkedSkills
-		if len(linkedSkills) > 0 {
-			r.setCondition(rt, ConditionTypeSkillsDiscovered, metav1.ConditionTrue, "SkillsFound",
-				fmt.Sprintf("%d linked skill(s) discovered from workload annotation", len(linkedSkills)))
-		} else {
-			meta.RemoveStatusCondition(&rt.Status.Conditions, ConditionTypeSkillsDiscovered)
-		}
 	} else {
 		rt.Status.LinkedSkills = nil
-		meta.RemoveStatusCondition(&rt.Status.Conditions, ConditionTypeSkillsDiscovered)
 	}
 	desired := rt.Status.DeepCopy()
 	if err := retry.RetryOnConflict(retry.DefaultRetry, func() error {
@@ -726,7 +714,6 @@ func (r *AgentRuntimeReconciler) handleDeletion(ctx context.Context, rt *agentv1
 			delete(workloadLabels, LabelManagedBy)
 			acc.obj.SetLabels(workloadLabels)
 
-
 			// Remove kagenti.io/type from PodTemplateSpec pod labels so future pods
 			// are not presented to the webhook with the type label.
 			podLabels := acc.getPodLabels(acc.obj)
@@ -770,10 +757,6 @@ func (r *AgentRuntimeReconciler) handleDeletion(ctx context.Context, rt *agentv1
 	return ctrl.Result{}, nil
 }
 
-func (r *AgentRuntimeReconciler) setPhase(rt *agentv1alpha1.AgentRuntime, phase agentv1alpha1.RuntimePhase) {
-	rt.Status.Phase = phase
-}
-
 func (r *AgentRuntimeReconciler) setCondition(rt *agentv1alpha1.AgentRuntime, condType string, status metav1.ConditionStatus, reason, message string) {
 	meta.SetStatusCondition(&rt.Status.Conditions, metav1.Condition{
 		Type:               condType,
@@ -784,16 +767,15 @@ func (r *AgentRuntimeReconciler) setCondition(rt *agentv1alpha1.AgentRuntime, co
 	})
 }
 
-// updateErrorStatus sets the AgentRuntime phase to Error and updates a condition
-// with retry-on-conflict semantics, re-fetching the object on each attempt.
+// updateErrorStatus updates a condition to False with retry-on-conflict
+// semantics, re-fetching the object on each attempt.
 func (r *AgentRuntimeReconciler) updateErrorStatus(ctx context.Context, key types.NamespacedName, condType, reason, message string) {
 	logger := log.FromContext(ctx)
 	if statusErr := retry.RetryOnConflict(retry.DefaultRetry, func() error {
 		latest := &agentv1alpha1.AgentRuntime{}
 		if err := r.Get(ctx, key, latest); err != nil {
 			return err
 		}
-		r.setPhase(latest, agentv1alpha1.RuntimePhaseError)
 		r.setCondition(latest, condType, metav1.ConditionFalse, reason, message)
 		return r.Status().Update(ctx, latest)
 	}); statusErr != nil {
@@ -810,8 +792,6 @@ func (r *AgentRuntimeReconciler) fetchAndUpdateCard(ctx context.Context, rt *age
 	if !r.EnableCardDiscovery {
 		if rt.Status.Card != nil {
 			rt.Status.Card = nil
-			r.setCondition(rt, ConditionTypeCardFetched, metav1.ConditionFalse, "DiscoveryDisabled",
-				"Card discovery is disabled; stale card data cleared")
 		}
 		return
 	}
@@ -823,29 +803,24 @@ func (r *AgentRuntimeReconciler) fetchAndUpdateCard(ctx context.Context, rt *age
 		lastHash = annotations[AnnotationLastCardFetchHash]
 	}
 	if changeKey != "" && changeKey == lastHash && rt.Status.Card != nil {
-		r.setCondition(rt, ConditionTypeCardFetched, metav1.ConditionTrue, "FetchSkipped",
-			"Pod template unchanged; existing card data still valid")
 		return
 	}
 
 	if ready, msg := r.checkWorkloadReady(ctx, rt.Namespace, rt.Spec.TargetRef); !ready {
 		logger.V(1).Info("Workload not ready for card discovery", "reason", msg)
-		r.setCondition(rt, ConditionTypeCardFetched, metav1.ConditionFalse, "WorkloadNotReady", msg)
 		return
 	}
 
 	svc, port, err := r.resolveServiceForWorkload(ctx, rt.Namespace, rt.Spec.TargetRef)
 	if err != nil {
 		logger.V(1).Info("Service resolution failed for card discovery", "error", err)
-		r.setCondition(rt, ConditionTypeCardFetched, metav1.ConditionFalse, "ServiceNotFound", err.Error())
 		return
 	}
 
 	protocol := agentcard.A2AProtocol
 	cardData, fetchResult, transportSecurity, err := r.fetchCard(ctx, rt, svc, port, protocol)
 	if err != nil {
 		logger.Error(err, "Card fetch failed", "workload", rt.Spec.TargetRef.Name)
-		r.setCondition(rt, ConditionTypeCardFetched, metav1.ConditionFalse, "FetchFailed", err.Error())
 		return
 	}
 
@@ -883,13 +858,6 @@ func (r *AgentRuntimeReconciler) fetchAndUpdateCard(ctx context.Context, rt *age
 
 	rt.Status.Card = cardStatus
 
-	conditionReason := "Fetched"
-	if transportSecurity == agentv1alpha1.TransportSecurityHTTP {
-		conditionReason = "FetchedInsecure"
-	}
-	r.setCondition(rt, ConditionTypeCardFetched, metav1.ConditionTrue, conditionReason,
-		fmt.Sprintf("Successfully fetched agent card for %s", cardData.Name))
-
 	r.persistCardFetchAnnotation(ctx, rt, changeKey)
 }