fix: apply CodeRabbit auto-fixes

Fixed 15 file(s) based on 20 unresolved review comments.

Co-authored-by: CodeRabbit <noreply@coderabbit.ai>

Author: coderabbitai[bot]

.agents/skills/agentrax-context/SKILL.md

@@ -3,6 +3,8 @@ name: agentrax-context
 description: Project context and settled architecture decisions for the Agentrax Kubernetes operator (module agentrax.io/v1alpha1, repo agentrax). Always consult this before writing, reviewing, or reasoning about any code in this repository — CRD types, the reconciler, the rollout controller, the autoscaler, the quota webhook, or the MCP registry — so implementation stays consistent with the design doc instead of drifting or re-deriving decisions that are already settled. Trigger on any mention of AgentDeployment, TenantQuota, canary rollout, or this repo's controllers, even if the user doesn't name the skill directly.
 ---
 
+# Agentrax Context Skill
+
 > When uncertain about any architecture decision, defer to `docs/ARCHITECTURE.md` rather than improvising. Don't guess when the doc has the answer.
 
 ## Non-negotiable terminology
@@ -26,7 +28,7 @@ description: Project context and settled architecture decisions for the Agentrax
 | Package                | Responsibility                                                                                                                   |
 | ---------------------- | -------------------------------------------------------------------------------------------------------------------------------- |
 | `api/v1alpha1/`        | CRD Go types, validation markers, defaulting. No business logic.                                                                 |
-| `internal/controller/` | Reconcile loops. Only code that calls the Kubernetes API for core owned resources.                                               |
+| `internal/controller/` | Reconcile loops. Only code that calls the Kubernetes API for core-owned resources.                                               |
 | `internal/rollout/`    | Canary state machine and PromQL threshold evaluation.                                                                            |
 | `internal/scaling/`    | HPA generation and quota-capped scaling logic.                                                                                   |
 | `internal/registry/`   | MCP registrar, registry HTTP handler, TTL sweep.                                                                                 |

README.md

@@ -94,12 +94,14 @@ flowchart TB
    cd agentrax
    ```
 
-2. **Install cluster dependencies** (cert-manager, Prometheus Operator, Gateway API CRDs):
+2. **Install cluster dependencies** (cert-manager, Prometheus Operator, Gateway API CRDs, and Prometheus Adapter):
 
    ```bash
    make deploy-deps
    ```
 
+   Note: Prometheus Adapter installation instructions are printed by `make deploy-deps`. Follow the displayed guidance to complete the metrics pipeline setup.
+
 3. **Install Agentrax CRDs:**
 
    ```bash
@@ -113,6 +115,7 @@ flowchart TB
    ```
 
 5. **Verify the operator is running:**
+
    ```bash
    kubectl get pods -n agentrax-system
    ```
@@ -245,14 +248,16 @@ curl -s http://localhost:9090/agents | jq .
 ]
 ```
 
+The `ttl` field is expressed in nanoseconds (e.g., `90000000000` = 90 seconds).
+
 ### Endpoints
 
-| Method   | Path                         | Description                                                       |
-| -------- | ---------------------------- | ----------------------------------------------------------------- |
-| `GET`    | `/agents`                    | List all active, non-expired registered agents.                   |
-| `GET`    | `/agents/{namespace}/{name}` | Get details and discovered tool capabilities of a specific agent. |
-| `POST`   | `/agents`                    | Register or update an agent entry directly.                       |
-| `DELETE` | `/agents/{namespace}/{name}` | Deregister an agent from the registry store.                      |
+| Method   | Path                         | Description                                                                                                                 |
+| -------- | ---------------------------- | --------------------------------------------------------------------------------------------------------------------------- |
+| `GET`    | `/agents`                    | List all active, non-expired registered agents.                                                                             |
+| `GET`    | `/agents/{namespace}/{name}` | Get details and discovered tool capabilities of a specific agent.                                                           |
+| `POST`   | `/agents`                    | Directly register or update an agent entry (bypasses MCP handshake; for administrative use or testing, not normal operation). |
+| `DELETE` | `/agents/{namespace}/{name}` | Deregister an agent from the registry store.                                                                                |
 
 ---
 

charts/agentrax/templates/deployment.yaml

@@ -58,27 +58,31 @@ spec:
               valueFrom:
                 fieldRef:
                   fieldPath: metadata.namespace
+            - name: AGENTRAX_REGISTRY_TTL
+              value: {{ .Values.registry.ttl | default "90s" | quote }}
+            - name: AGENTRAX_MCP_HEALTH_INTERVAL
+              value: {{ .Values.mcp.healthInterval | default "30s" | quote }}
             {{- range $key, $val := .Values.env }}
             - name: {{ $key }}
               value: {{ $val | quote }}
             {{- end }}
           ports:
             - name: registry
-              containerPort: 9090
+              containerPort: {{ .Values.registry.bindPort | default 9090 }}
               protocol: TCP
             - name: health
-              containerPort: 8081
+              containerPort: {{ .Values.manager.healthProbeBindPort | default 8081 }}
               protocol: TCP
           livenessProbe:
             httpGet:
               path: /healthz
-              port: 8081
+              port: {{ .Values.manager.healthProbeBindPort | default 8081 }}
             initialDelaySeconds: 15
             periodSeconds: 20
           readinessProbe:
             httpGet:
               path: /readyz
-              port: 8081
+              port: {{ .Values.manager.healthProbeBindPort | default 8081 }}
             initialDelaySeconds: 5
             periodSeconds: 10
           resources:

charts/agentrax/templates/registry-service.yaml

@@ -13,5 +13,5 @@ spec:
   ports:
     - name: registry
       port: {{ .Values.registry.service.port | default 9090 }}
-      targetPort: 9090
+      targetPort: {{ .Values.registry.bindPort | default 9090 }}
       protocol: TCP

charts/agentrax/values.yaml

@@ -68,6 +68,8 @@ manager:
   metricsBindAddress: "0"
   # -- Bind address for health/readiness probes.
   healthProbeBindAddress: ":8081"
+  # -- Bind port for health/readiness probes (referenced by liveness/readiness probes).
+  healthProbeBindPort: 8081
   # -- GPU resource name used for quota calculation.
   gpuResourceName: "nvidia.com/gpu"
 
@@ -94,6 +96,8 @@ registry:
     type: ClusterIP
   # -- Bind address for the in-operator registry HTTP server
   bindAddress: ":9090"
+  # -- Bind port for the registry HTTP server (referenced by container port and targetPort).
+  bindPort: 9090
   # -- Registry entry TTL (e.g. "90s")
   ttl: "90s"
 
@@ -103,6 +107,5 @@ mcp:
   healthInterval: "30s"
 
 # -- Environment variables injected into the manager container.
-env:
-  AGENTRAX_MCP_HEALTH_INTERVAL: "30s"
-  AGENTRAX_REGISTRY_TTL: "90s"
+# These values are automatically populated from registry.ttl and mcp.healthInterval above.
+env: {}

cmd/main.go

@@ -217,6 +217,24 @@ func main() {
 		registryNamespace = "agentrax-system"
 	}
 
+	registryTTL := registry.DefaultTTL
+	if v := os.Getenv("AGENTRAX_REGISTRY_TTL"); v != "" {
+		if parsed, err := time.ParseDuration(v); err == nil {
+			registryTTL = parsed
+		} else {
+			setupLog.Error(err, "invalid AGENTRAX_REGISTRY_TTL, using default", "default", registry.DefaultTTL)
+		}
+	}
+
+	mcpHealthInterval := 60 * time.Second
+	if v := os.Getenv("AGENTRAX_MCP_HEALTH_INTERVAL"); v != "" {
+		if parsed, err := time.ParseDuration(v); err == nil {
+			mcpHealthInterval = parsed
+		} else {
+			setupLog.Error(err, "invalid AGENTRAX_MCP_HEALTH_INTERVAL, using default", "default", mcpHealthInterval)
+		}
+	}
+
 	mgr, err := ctrl.NewManager(ctrl.GetConfigOrDie(), ctrl.Options{
 		Scheme:                 scheme,
 		Metrics:                metricsServerOptions,
@@ -272,7 +290,7 @@ func main() {
 	}
 
 	// Initialize MCP discovery registry and registrar.
-	mcpRegistry := registry.NewRegistry(mgr.GetClient(), registryNamespace, registry.DefaultTTL)
+	mcpRegistry := registry.NewRegistry(mgr.GetClient(), registryNamespace, registryTTL)
 	mcpRegistrar := registry.NewRegistrar(mcpRegistry, registry.NewHTTPMCPClient())
 
 	if registryAddr != "" && registryAddr != "0" {
@@ -287,11 +305,12 @@ func main() {
 	}
 
 	agentDeploymentReconciler := &controller.AgentDeploymentReconciler{
-		Client:           mgr.GetClient(),
-		Scheme:           mgr.GetScheme(),
-		GPUResourceName:  gpuResourceName,
-		CanaryController: canaryController,
-		Registrar:        mcpRegistrar,
+		Client:            mgr.GetClient(),
+		Scheme:            mgr.GetScheme(),
+		GPUResourceName:   gpuResourceName,
+		CanaryController:  canaryController,
+		Registrar:         mcpRegistrar,
+		MCPHealthInterval: mcpHealthInterval,
 	}
 	if canaryController != nil {
 		canaryController.Registrar = mcpRegistrar

config/manager/kustomization.yaml

@@ -5,5 +5,5 @@ apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 images:
 - name: controller
-  newName: example.com/agentrax
-  newTag: v0.1.0
+  newName: ghcr.io/gitcommitankit/agentrax
+  newTag: latest

docs/ARCHITECTURE.md

@@ -89,16 +89,16 @@ flowchart TB
 
 The repository enforces strict directional boundaries to prevent circular dependencies and isolate business logic from Kubernetes plumbing:
 
-| Package                                                                                         | Scope & Responsibility                                                                          | Key Invariants                                                                                       |
-| ----------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------- |
-| [`api/v1alpha1/`](file:///home/ankit/workspace/my-projects/agentrax/api/v1alpha1)               | CRD type definitions, OpenAPI markers, schema validation rules, and status condition constants. | **Zero business logic**; only struct declarations and generated deep-copy methods.                   |
-| [`internal/controller/`](file:///home/ankit/workspace/my-projects/agentrax/internal/controller) | Controller-runtime reconcile loops (`AgentDeployment`, `TenantQuota`).                          | Only layer that executes write calls against the Kubernetes API. Consumes subsystems via interfaces. |
-| [`internal/quota/`](file:///home/ankit/workspace/my-projects/agentrax/internal/quota)           | Quota arithmetic and concurrency-safe in-flight reservation cache.                              | Pure arithmetic; mutex-guarded state map; zero direct API server network calls in calculation paths. |
-| [`internal/webhook/`](file:///home/ankit/workspace/my-projects/agentrax/internal/webhook)       | Validating and Mutating admission webhooks.                                                     | Shared with `internal/quota` to enforce admission rules before objects are persisted.                |
-| [`internal/scaling/`](file:///home/ankit/workspace/my-projects/agentrax/internal/scaling)       | HPA synthesis, velocity rules, and dynamic quota ceiling headroom.                              | Calculates `QuotaHeadroom()` to cap HPA `maxReplicas` and applies stabilization windows.             |
-| [`internal/rollout/`](file:///home/ankit/workspace/my-projects/agentrax/internal/rollout)       | Canary state machine, PromQL query construction, and threshold evaluation.                      | Re-entrant state machine; sample-size gating; fail-safe timeout evaluation.                          |
-| [`internal/registry/`](file:///home/ankit/workspace/my-projects/agentrax/internal/registry)     | MCP registrar, JSON-RPC 2.0 handshake, TTL sweeper, and discovery REST API.                     | In-memory registry with ConfigMap write-through; background health probes and TTL sweep.             |
-| [`internal/metrics/`](file:///home/ankit/workspace/my-projects/agentrax/internal/metrics)       | Bounded HTTP Prometheus query client.                                                           | Wraps all responses with `io.LimitReader` (1 MiB ceiling) to prevent memory exhaustion.              |
+| Package                      | Scope & Responsibility                                                                          | Key Invariants                                                                                       |
+| ---------------------------- | ----------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------- |
+| `api/v1alpha1/`              | CRD type definitions, OpenAPI markers, schema validation rules, and status condition constants. | **Zero business logic**; only struct declarations and generated deep-copy methods.                   |
+| `internal/controller/`       | Controller-runtime reconcile loops (`AgentDeployment`, `TenantQuota`).                          | Only layer that executes write calls against the Kubernetes API for core-owned resources (Deployments, Services, HPAs, HTTPRoutes). Consumes subsystems via interfaces. |
+| `internal/quota/`            | Quota arithmetic and concurrency-safe in-flight reservation cache.                              | Pure arithmetic; mutex-guarded state map; zero direct API server network calls in calculation paths. |
+| `internal/webhook/`          | Validating and Mutating admission webhooks.                                                     | Shared with `internal/quota` to enforce admission rules before objects are persisted.                |
+| `internal/scaling/`          | HPA synthesis, velocity rules, and dynamic quota ceiling headroom.                              | Calculates `QuotaHeadroom()` to cap HPA `maxReplicas` and applies stabilization windows.             |
+| `internal/rollout/`          | Canary state machine, PromQL query construction, and threshold evaluation.                      | Re-entrant state machine; sample-size gating; fail-safe timeout evaluation.                          |
+| `internal/registry/`         | MCP registrar, JSON-RPC 2.0 handshake, TTL sweeper, and discovery REST API.                     | In-memory registry with ConfigMap write-through for persistence; background health probes and TTL sweep. Explicitly allowed to write the `agentrax-registry` ConfigMap for state recovery. |
+| `internal/metrics/`          | Bounded HTTP Prometheus query client.                                                           | Wraps all responses with `io.LimitReader` (1 MiB ceiling) to prevent memory exhaustion.              |
 
 ---
 

internal/controller/agentdeployment_controller.go

@@ -97,6 +97,11 @@ type AgentDeploymentReconciler struct {
 	// When nil, MCP registration features are disabled.
 	Registrar AgentRegistrar
 
+	// MCPHealthInterval is the requeue interval used to drive periodic MCP
+	// heartbeat probes. Must stay shorter than the registry TTL so heartbeats
+	// land before entries expire. Defaults to 60s when zero.
+	MCPHealthInterval time.Duration
+
 	// hasServiceMonitorCRD is set once during SetupWithManager and determines
 	// whether ServiceMonitor reconciliation is attempted at all.
 	hasServiceMonitorCRD bool
@@ -618,8 +623,11 @@ func (r *AgentDeploymentReconciler) reconcileMCPRegistration(ctx context.Context
 		return 0
 	}
 
-	// Requeue interval shorter than TTL (90s) to ensure heartbeats occur before expiry.
-	requeueInterval := 60 * time.Second
+	// Requeue interval shorter than the registry TTL to ensure heartbeats occur before expiry.
+	requeueInterval := r.MCPHealthInterval
+	if requeueInterval <= 0 {
+		requeueInterval = 60 * time.Second
+	}
 
 	// If already registered, perform heartbeat probe.
 	if ad.Status.Registered {

internal/controller/agentdeployment_controller_test.go

@@ -208,15 +208,15 @@ var _ = Describe("AgentDeployment Controller", func() {
 			// (called on the reconciler goroutine) can pass its observation to the
 			// test goroutine without a data race on plain booleans.
 			resultCh := make(chan bool, 1)
-			origRegistrar := testReconciler.Registrar
-			testReconciler.Registrar = &mockAgentRegistrar{
+			mockReg := &mockAgentRegistrar{
 				deregisterFn: func(hctx context.Context, had *agentraxv1alpha1.AgentDeployment) error {
 					err := k8sClient.Get(hctx, key, &corev1.Service{})
 					resultCh <- (err == nil)
 					return nil
 				},
 			}
-			DeferCleanup(func() { testReconciler.Registrar = origRegistrar })
+			testRegistrarProxy.SetDelegate(mockReg)
+			DeferCleanup(func() { testRegistrarProxy.SetDelegate(testRegistrar) })
 
 			// Delete the object — the reconciler must call Deregister, then remove the finalizer.
 			Expect(k8sClient.Delete(ctx, ad)).To(Succeed())