feat: add comprehensive Kubernetes agent skill and k8s-cli

jasper commented

2026-07-11 15:42:37 -04:00

Contributor

Summary

Add a broad, deeply researched kubernetes Agent Skills skill to this repository. The skill should help agents safely design, deploy, operate, troubleshoot, secure, upgrade, and automate Kubernetes clusters and workloads across major distributions and managed targets, with first-class coverage for k3s.

Scope

Upstream Kubernetes and kubeadm operations
k3s and RKE2
MicroK8s, k0s, Talos, kind, and Minikube where current source coverage supports useful guidance
OpenShift/OKD and Rancher-managed environments with platform-specific boundaries
Major managed Kubernetes providers confirmed during research
Kubernetes API/resource workflows, workloads, networking, storage, scheduling, autoscaling, security, observability, upgrades, backup/restore, and troubleshooting
Helm, Kustomize, GitOps, server-side apply, discovery, dry-run, rollout verification, and safe mutation patterns
A tested, agent-first scripts/k8s-cli wrapper with structured output, bounded diagnostics, safety gates, and a native-tool escape hatch

Research Requirements

Crawl complete official documentation sites before selecting individual reference pages.
Research intermediate and advanced tutorials, production practices, troubleshooting guides, official forums, release notes, and issue discussions.
Build and preserve a recursive artifact pyramid before authoring the skill.
Record source URLs, versions, publication/update dates, last-reviewed dates, and version-sensitive claims.
Reconcile secondary material against current primary documentation and live CLI help where possible.

Deliverables

kubernetes/SKILL.md, compliant with agentskills.io progressive disclosure and size guidance
Human-facing kubernetes/README.md
Source index, version matrix, API validation audit, and research gap audit
Focused references for core operations, distributions/providers, security, reliability, troubleshooting, and delivery tooling
Useful templates and assets only where they have stable operational value
Executable scripts/k8s-cli plus deterministic and live smoke-test coverage where a disposable cluster is available
Repository index/routing updates required by existing conventions
Greenfield SkillOpt review and full contribution-pipeline delivery

Acceptance Criteria

Broad research artifact pyramid exists before skill authoring
Every declared coverage area is source-backed or explicitly documented as a gap/scope boundary
Version-sensitive API, command, provider, and distribution claims have primary-source validation
k3s receives first-class installation, architecture, lifecycle, upgrade, and troubleshooting coverage
k8s-cli is non-interactive, agent-friendly, structured, bounded, safe by default, and documented
Agent Skills structural validation passes
Repository validators and README checks pass
Security/privacy audit finds no private infrastructure data or unsafe secrets
CI is green
PR is reviewed and merged

Filed by Jasper (AI agent on behalf of Magnus Hedemark)

## Summary Add a broad, deeply researched `kubernetes` Agent Skills skill to this repository. The skill should help agents safely design, deploy, operate, troubleshoot, secure, upgrade, and automate Kubernetes clusters and workloads across major distributions and managed targets, with first-class coverage for k3s. ## Scope - Upstream Kubernetes and kubeadm operations - k3s and RKE2 - MicroK8s, k0s, Talos, kind, and Minikube where current source coverage supports useful guidance - OpenShift/OKD and Rancher-managed environments with platform-specific boundaries - Major managed Kubernetes providers confirmed during research - Kubernetes API/resource workflows, workloads, networking, storage, scheduling, autoscaling, security, observability, upgrades, backup/restore, and troubleshooting - Helm, Kustomize, GitOps, server-side apply, discovery, dry-run, rollout verification, and safe mutation patterns - A tested, agent-first `scripts/k8s-cli` wrapper with structured output, bounded diagnostics, safety gates, and a native-tool escape hatch ## Research Requirements - Crawl complete official documentation sites before selecting individual reference pages. - Research intermediate and advanced tutorials, production practices, troubleshooting guides, official forums, release notes, and issue discussions. - Build and preserve a recursive artifact pyramid before authoring the skill. - Record source URLs, versions, publication/update dates, last-reviewed dates, and version-sensitive claims. - Reconcile secondary material against current primary documentation and live CLI help where possible. ## Deliverables - `kubernetes/SKILL.md`, compliant with agentskills.io progressive disclosure and size guidance - Human-facing `kubernetes/README.md` - Source index, version matrix, API validation audit, and research gap audit - Focused references for core operations, distributions/providers, security, reliability, troubleshooting, and delivery tooling - Useful templates and assets only where they have stable operational value - Executable `scripts/k8s-cli` plus deterministic and live smoke-test coverage where a disposable cluster is available - Repository index/routing updates required by existing conventions - Greenfield SkillOpt review and full contribution-pipeline delivery ## Acceptance Criteria - [ ] Broad research artifact pyramid exists before skill authoring - [ ] Every declared coverage area is source-backed or explicitly documented as a gap/scope boundary - [ ] Version-sensitive API, command, provider, and distribution claims have primary-source validation - [ ] k3s receives first-class installation, architecture, lifecycle, upgrade, and troubleshooting coverage - [ ] `k8s-cli` is non-interactive, agent-friendly, structured, bounded, safe by default, and documented - [ ] Agent Skills structural validation passes - [ ] Repository validators and README checks pass - [ ] Security/privacy audit finds no private infrastructure data or unsafe secrets - [ ] CI is green - [ ] PR is reviewed and merged Filed by Jasper (AI agent on behalf of Magnus Hedemark)

jasper referenced this issue

2026-07-11 16:27:42 -04:00

feat: add comprehensive Kubernetes agent skill #103

jasper commented

2026-07-11 16:49:44 -04:00

Author

Contributor

Resolved by merged PR #103: #103

Filed by Jasper (AI agent on behalf of Magnus Hedemark)

Resolved by merged PR #103: https://git.brandyapple.com/magnus/agent-skills/pulls/103 Filed by Jasper (AI agent on behalf of Magnus Hedemark)

jasper closed this issue

2026-07-11 16:49:45 -04:00