fix(cli): make agent autoscaling opt-in instead of a hardcoded default

A crash-looping pod's restart CPU burn reads as CPU load, so the HPA this
CLI attached to every deployment scaled the failure to maxReplicas, then
froze there once the metrics went missing.

Keep the autoscaling block so an environments.yaml opt-in still deep-merges
onto the existing replica bounds and CPU target.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

Author: mohammadatallah-scale

src/agentex/lib/cli/handlers/deploy_handlers.py

@@ -292,9 +292,9 @@ def merge_deployment_configs(
                 "memory": manifest.deployment.global_config.resources.limits.memory,
             },
         },
-        # Enable autoscaling by default for production deployments
+        # Opt-in only: a crash loop's restart CPU burn reads as load and scales it to max.
         "autoscaling": {
-            "enabled": True,
+            "enabled": False,
             "minReplicas": 1,
             "maxReplicas": 10,
             "targetCPUUtilizationPercentage": 50,
@@ -307,9 +307,8 @@ def merge_deployment_configs(
         if temporal_config:
             helm_values[TEMPORAL_WORKER_KEY] = {
                 "enabled": True,
-                # Enable autoscaling for temporal workers as well
                 "autoscaling": {
-                    "enabled": True,
+                    "enabled": False,
                     "minReplicas": 1,
                     "maxReplicas": 10,
                     "targetCPUUtilizationPercentage": 50,

src/agentex/lib/cli/templates/default-claude-code/environments.yaml.j2

@@ -51,6 +51,9 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true
       temporal:
         enabled: false
 

src/agentex/lib/cli/templates/default-codex/environments.yaml.j2

@@ -51,6 +51,9 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true
       temporal:
         enabled: false
 

src/agentex/lib/cli/templates/default-langgraph/environments.yaml.j2

@@ -51,6 +51,9 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true
       temporal:
         enabled: false
 

src/agentex/lib/cli/templates/default-openai-agents/environments.yaml.j2

@@ -50,4 +50,7 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true
 

src/agentex/lib/cli/templates/default-pydantic-ai/environments.yaml.j2

@@ -51,6 +51,9 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true
       temporal:
         enabled: false
 

src/agentex/lib/cli/templates/default/environments.yaml.j2

@@ -51,6 +51,9 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true
       temporal:
         enabled: false
 

src/agentex/lib/cli/templates/sync-claude-code/environments.yaml.j2

@@ -50,4 +50,7 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true
 

src/agentex/lib/cli/templates/sync-codex/environments.yaml.j2

@@ -50,4 +50,7 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true
 

src/agentex/lib/cli/templates/sync-langgraph/environments.yaml.j2

@@ -50,4 +50,7 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
+      # Autoscaling is opt-in. Unset keys inherit maxReplicas 10 and a 50% CPU target.
+      # autoscaling:
+      #   enabled: true