kubernetes-architect

What This Skill Does

The kubernetes-architect skill is a specialized AI agent designed to act as a senior platform engineer. It focuses on the generation of robust, production-ready Kubernetes manifests. Unlike generic AI code generators, this skill is purpose-built to adhere to GitOps methodologies, favoring declarative configuration for tools like ArgoCD and Flux. It understands the nuances of various managed cloud providers, including EKS, AKS, and GKE, and can configure sophisticated service mesh integrations such as Istio or Linkerd to ensure traffic security, observability, and advanced deployment patterns like canary or blue-green releases.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/anton-abyzov/sw-kubernetes-architect

Use Cases

This skill is ideal for DevOps engineers and SREs looking to scale their infrastructure without technical debt. Common use cases include:

• Standardizing ingress controllers and service mesh configurations across clusters.
• Designing high-availability deployments with proper resource requests and limits.
• Transitioning legacy Helm charts into structured GitOps repositories.
• Implementing secure Kubernetes Network Policies to enforce zero-trust networking.

Example Prompts

1. "Generate a Kubernetes deployment manifest for a Node.js microservice that includes a readiness probe, resource limits, and an Istio virtual service for canary routing."
2. "Create a sealed-secret template compatible with ArgoCD that injects database credentials as environment variables for my staging environment."
3. "Refactor this existing Deployment manifest to support HPA (Horizontal Pod Autoscaler) and add affinity rules to ensure pods are spread across multiple availability zones in EKS."

Tips & Limitations

• Granularity: Always prompt for one service at a time to ensure the generated manifests remain focused and modular. This prevents large, error-prone configurations that can crash clusters.
• Context: Provide details about your specific cloud provider (e.g., EKS or GKE) in the prompt to allow the agent to optimize IAM roles or load balancer annotations.
• Validation: While the code is high-quality, always pass the generated output through kube-linter or polaris before applying it to a production cluster to verify against your specific security policies.