fabrico-deploy-kubernetes

name	fabrico-deploy-kubernetes
description	Create production-ready Kubernetes and Helm configuration.

Invocation portability: $fabrico-* below means the discovered entry workflow. Use the unqualified name for repository or filesystem installs and $fabrico-collections:fabrico-* for plugin installs.

Input

Use the user’s current request as the workflow input. Expected context: [service or workload to deploy].

If a named custom agent is unavailable, as in a skills-only plugin installation, perform that delegated step in the current thread with the referenced skills and the same safety and verification gates.

Kubernetes Deployment Workflow

This workflow creates Kubernetes deployments, Helm charts, and workload configurations following production-ready patterns for high availability, resource management, and secrets handling. It ensures consistent deployment practices across environments with proper health probes, scaling policies, and security configurations.

The workflow matches existing project conventions (raw manifests, Helm, or Kustomize), configures appropriate QoS classes and resource limits, and validates all manifests before deployment. Architectural decisions involving service mesh, multi-cluster, or microservices topology are escalated to the architect subagent.

Required Skills

Before starting, load and follow these skills:

• fabrico-implementing-kubernetes - for deployment patterns, Helm charts, resource configuration, and high availability settings
• fabrico-managing-secrets - for Kubernetes secrets management (Secrets, external-secrets, sealed-secrets)
• fabrico-technical-context-discovering - to establish project conventions and existing Kubernetes patterns

---

1. Context

Follow the fabrico-technical-context-discovering skill to identify existing Kubernetes setup.

Additionally, always:

• Check root and nested AGENTS.md guidance → project-specific conventions
• Analyze existing Kubernetes manifests, Helm charts, or Kustomize overlays
• Discover namespace organization and resource patterns

---

2. Implementation

Follow the fabrico-implementing-kubernetes skill for:

• Deployment/StatefulSet configuration
• Service and Ingress setup
• Resource requests/limits and QoS classes
• Health probes (liveness, readiness, startup)
• HPA, PDB, pod anti-affinity, topology spread

Follow the fabrico-managing-secrets skill for:

• Kubernetes Secrets or external secrets management
• Secret injection configuration

---

3. Architect Consultation

Request fabrico-architect consultation when:

• Designing new service architecture or microservices topology
• Implementing service mesh (Istio, Linkerd)
• Setting up multi-cluster or multi-region deployments

Skip for: HPA/PDB configuration, probes, resource limits, manifest fixes.

When already running inside fabrico-devops-engineer, return a focused architecture-decision request to the caller; do not spawn a sibling or parent. Otherwise delegate to the architect when available, or perform the consultation in the current thread with fabrico-architecture-designing.

---

4. Summary (required output)

## Kubernetes Deployment Summary

### Current State
- [existing Kubernetes configuration]

### Proposed Configuration
- Deployment method: [raw manifests / Helm / Kustomize]
- Namespace: [target namespace]
- Replicas: [count] with HPA [min-max]

### Prerequisites
| Resource | Description |
|----------|-------------|
| [namespace/secret/etc.] | [what must exist before deployment] |

### Deployment Instructions
1. [step-by-step instructions to apply]

### Verification Steps
1. [how to confirm successful deployment]

### Files
- NEW/MODIFIED: [list of files created or modified]

---

Scope

Does NOT handle (redirect to):

• Cluster provisioning → $fabrico-implement-terraform
• CI/CD pipeline for deployment → $fabrico-implement-pipeline
• Monitoring and alerting → $fabrico-implement-observability