Zelkor

Production Install ​

Deploy the production shape of Zelkor Community Edition. This runs databases via Kubernetes operators (CloudNativePG, ClickHouse Operator) and enables High Availability (HA) and NetworkPolicies. The workload is a normal Helm release.

Zelkor sandboxes the agent you already wrote. It wraps the agent in a comprehensive security and operational perimeter without a rewrite. The agent can't break out, reach unauthorized data or networks, its prompts are verified, budget is controlled, and it is under observation.

  • Community Edition is the self-hosted runtime. It can run this highly available operator shape out of the box.
  • Pro adds SSO, team controls (budgets and approvals), and team GitOps on top of this shape.
  • Enterprise adds isolation and compliance on Pro (hardware sandbox, mTLS, retained audit, BAA).

Prerequisites ​

  • Kubernetes v1.28+. The production profile runs 3 Postgres instances. The installer stops when the Postgres StorageClass is not on that many Ready nodes and tells you the --set to pass.
  • A default StorageClass that can provision a volume, or all three class keys: databases.postgresql.storage.storageClass, databases.clickhouse.storage.storageClass, seaweedfs.persistence.storageClass.
  • metrics-server installed (required for HPA).
  • A JWT Identity Provider (IdP) for tenant authentication, with a downloaded JWKS JSON file (or an internal OIDC issuer like Keycloak for isolated environments).

NetworkPolicies are disabled by default. Set security.networkPolicies.enabled: true in your values to enforce Network Boundaries and isolate pod traffic.

Sandbox Runtime (gVisor) ​

Community Edition uses gVisor to sandbox generated code. On production clusters, you must install the runsc binary on your worker nodes and configure a Kubernetes RuntimeClass named gvisor. (Managed Kubernetes services like GKE and AKS often provide this as a built-in node pool option). If the RuntimeClass is missing, sandbox worker pods will fail to schedule.

Deploy with the production script ​

The install-production.sh script bootstraps the required operators, installs Envoy Gateway, and deploys the platform using the values-production.yaml profile.

Replace placeholders with your actual hosts, keys, and JWT settings ​

bash
git clone https://github.com/devopssquaddev/zelkor-platform.git
cd zelkor-platform

OPENAI_API_KEY=sk-... ./scripts/install-production.sh \
  --namespace zelkor \
  --hosts-agents agents.example.com \
  --hosts-langfuse langfuse.example.com \
  --jwt-issuer "https://your-idp.example.com" \
  --jwt-audience "zelkor" \
  --jwks-file "./path/to/jwks.json" \
  --generate-passwords

This script maps the JWT flags to the underlying Helm keys (platform.tenants.jwt.issuer, audiences[0], and jwksConfigMap).

Note on JWT Issuers: Production installs strictly enforce identity verification. If you do not have an external IdP (like Auth0 or Entra ID) and are deploying to an isolated or air-gapped environment, you have two alternatives:

  1. Run a lightweight internal OIDC provider (like Keycloak or Zitadel) in your cluster and point --jwt-issuer to it.
  2. Use Zelkor's native dev signing (not recommended for true production). To bypass the issuer requirement, pass the localSigning override instead of the --jwt-* flags:
bash
OPENAI_API_KEY=sk-... ./scripts/install-production.sh \
  --hosts-agents agents.example.com \
  --hosts-langfuse langfuse.example.com \
  --set "platform.tenants.jwt.localSigning.enabled=true" \
  --generate-passwords

Store the generated passwords (such as POSTGRES_PASSWORD) securely. The script will configure Envoy Gateway using a standard LoadBalancer by default. If you have an existing Ingress controller, you can use the --topology layered option.

Optional configurations ​

  • TLS: Use --tls --cluster-issuer letsencrypt-prod to attach a cert-manager ClusterIssuer to the Gateway.
  • ServiceMonitor: Use --service-monitor to enable Prometheus metrics scraping.
  • Existing operators: Pass --skip-operators when CloudNativePG, ClickHouse Operator, or cert-manager are already on the cluster.
  • Storage: Leave the three volume classes empty only when the default StorageClass can provision. If you set one of databases.postgresql.storage.storageClass, databases.clickhouse.storage.storageClass, or seaweedfs.persistence.storageClass, set all three. Postgres stays at 3 instances unless the installer prints --set databases.postgresql.instances=N.
  • gVisor: Set security.sandbox.nodes.selector to nodes that have RuntimeClass gvisor.
  • Layered edge: --topology layered prints the Envoy dataplane Service and an Ingress example (namespace envoy-gateway-system, preserve Host). The script does not apply it. Health checks are https://<agents-host>/health and https://<langfuse-host>/api/public/health. Do not wrap the dataplane in another ClusterIP Endpoints list.
  • LLM keys on upgrade: keep passing --set-file for workspace.models.providers.*.apiKey, or omit the key. An overlay with apiKey: "" deletes the AI Gateway route (route not found).

--jwt-issuer must match the token iss. Prefer --jwks-file. Remote JWKS needs HTTPS plus jwksEgressCIDRs when NetworkPolicies are on (JWT rejected).

Next steps ​