Helm Values
All configurable values for the arkonis/ark-operator Helm chart.
helm repo add arkonis https://charts.arkonis.dev
helm show values arkonis/ark-operator
Core configuration
| Value | Default | Description |
taskQueueURL | "" | Redis connection string for the task queue. Required. Example: redis.ark-system.svc.cluster.local:6379 |
triggerWebhook.url | "" | Base URL of the webhook receiver exposed by the operator. Example: http://ark-operator.ark-system.svc.cluster.local:8092 |
replicaCount | 1 | Number of operator pod replicas. |
namespace | ark-system | Namespace where the operator is deployed. |
API keys
| Value | Default | Description |
apiKeys.openaiApiKey | "" | OpenAI API key. Injected into all agent pods as OPENAI_API_KEY. |
apiKeys.anthropicApiKey | "" | Anthropic API key. Injected into all agent pods as ANTHROPIC_API_KEY. |
apiKeys.existingSecret | "" | Name of an existing Secret containing API key env vars. If set, apiKeys.openaiApiKey and apiKeys.anthropicApiKey are ignored. |
Agent pod configuration
| Value | Default | Description |
agentImage | ghcr.io/arkonis-dev/ark-runtime:latest | Container image for agent pods. |
agentImagePullPolicy | IfNotPresent | Image pull policy for agent pods. |
agentExtraEnv | [] | Additional environment variables injected into all agent pods. Useful for cluster-wide provider config (e.g. OPENAI_BASE_URL, AGENT_PROVIDER). |
agentResources.requests.cpu | 100m | CPU request for agent pods. |
agentResources.requests.memory | 256Mi | Memory request for agent pods. |
agentResources.limits.cpu | 500m | CPU limit for agent pods. |
agentResources.limits.memory | 512Mi | Memory limit for agent pods. |
agentExtraEnv:
- name: AGENT_PROVIDER
value: openai
- name: OPENAI_BASE_URL
value: http://ollama.ollama.svc.cluster.local:11434/v1
- name: OPENAI_API_KEY
value: ollama
Operator pod configuration
| Value | Default | Description |
operatorImage | ghcr.io/arkonis-dev/ark-operator:latest | Container image for the operator. |
operatorImagePullPolicy | IfNotPresent | Image pull policy for the operator. |
resources.requests.cpu | 100m | CPU request for the operator pod. |
resources.requests.memory | 128Mi | Memory request for the operator pod. |
resources.limits.cpu | 500m | CPU limit for the operator pod. |
resources.limits.memory | 256Mi | Memory limit for the operator pod. |
Observability
| Value | Default | Description |
otel.endpoint | "" | OTel collector endpoint. When set, forwarded to the operator and all agent pods as OTEL_EXPORTER_OTLP_ENDPOINT. |
otel.serviceName | ark-operator | Service name reported in traces from the operator. Agent pods use ark-agent. |
Image pull secrets
| Value | Default | Description |
imagePullSecrets | [] | List of image pull secret names for both operator and agent pods. |
Example: Ollama installation
helm install ark-operator arkonis/ark-operator \
--namespace ark-system \
--create-namespace \
--set taskQueueURL=redis.ark-system.svc.cluster.local:6379 \
--set triggerWebhook.url=http://ark-operator.ark-system.svc.cluster.local:8092 \
--set agentExtraEnv[0].name=AGENT_PROVIDER,agentExtraEnv[0].value=openai \
--set agentExtraEnv[1].name=OPENAI_BASE_URL,agentExtraEnv[1].value=http://ollama.ollama.svc.cluster.local:11434/v1 \
--set agentExtraEnv[2].name=OPENAI_API_KEY,agentExtraEnv[2].value=ollama
Example: OpenAI with existing secret
kubectl create secret generic ark-api-keys \
--from-literal=OPENAI_API_KEY=sk-... \
--namespace ark-system
helm install ark-operator arkonis/ark-operator \
--namespace ark-system \
--create-namespace \
--set taskQueueURL=redis.ark-system.svc.cluster.local:6379 \
--set apiKeys.existingSecret=ark-api-keys
See also