Add kubernates skill rule

[DeepsanBhandari]

Summary

Adds a new Kubernetes skill rule providing best practices for container orchestration, including deployment patterns, Helm usage, kubectl workflows, and resource management.

Closes #2

Type

• New rule (config/rules/, config/commands/, config/skills/, etc.)
• Rule improvement (existing rule)
• CLI feature / fix (src/)
• Bug fix
• Documentation

Changes

• config/skills/kubernetes.md

  • Added Kubernetes skill rule with deployment patterns, resource limits, health checks, Helm best practices, and kubectl workflows.

Why

This rule provides guidance for deploying and operating applications on Kubernetes clusters. It helps standardize best practices for manifests, resource management, and operational workflows when working with containerized applications.

Validation

• npx vitest run ✅
• npm run build ✅

For Rule PRs

• Has description field in frontmatter
• Under 100 lines
• No overlap with existing rules
• Tested with npx ai-nexus test "deploy to kubernetes"

[DeepsanBhandari]

PR submitted for the kubernetes skill rule. Feedback is welcome. Thanks!

[JSK9999]

Thanks for the contribution! The content covers what #2 asked for nicely. Just a few things to adjust before merging.

1. Line count (146 lines)

The guideline in CONTRIBUTING.md asks for under 100 lines, and this one's at 146. If you look at docker.md, it's about 50 lines for a similar scope. I think trimming the YAML examples would get you there — keeping one Deployment example and
converting the rest to bullet points should be enough.

2. keywords in frontmatter

Other skill files like docker.md have a keywords field for routing. Would be good to add one here too:

---
description: Kubernetes and container orchestration best practices
keywords: [kubernetes, k8s, kubectl, helm, pod, deployment]
---

3. Tone — more prescriptive, less reference-y

Some sections read a bit like docs rather than guidance. The kubectl Workflows section for example just lists commands without much context on when to reach for what. If you look at how docker.md is written — "Use multi-stage builds to minimize image size", "Don't run as root" — each line gives a clear direction. Matching that style would keep things consistent across the skill files.