k8s

Operate the joelclaw Kubernetes cluster — Talos Linux on Colima (Mac Mini). Deploy services, check health, debug pods, recover from restarts, add ports, manage Helm releases, inspect logs, fix networking. Triggers on: 'kubectl', 'pods', 'deploy to k8s', 'cluster health', 'restart pod', 'helm install', 'talosctl', 'colima', 'nodeport', 'flannel', 'port mapping', 'k8s down', 'cluster not working', 'add a port', 'PVC', 'storage', any k8s/Talos/Colima infrastructure task. Also triggers on service-specific deploy: 'deploy redis', 'redeploy inngest', 'livekit helm', 'pds not responding'.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Medium

Suggest reviewing before use

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An exceptionally actionable, well-structured operations skill with concrete commands, complete manifests, and validation-aware recovery workflows. Its main weakness is conciseness: a large body with narrative flavor and inlined detail that could be trimmed or moved to the single reference file.

Suggestions

Trim narrative flavor and lengthy incident rationales (e.g. 'not healer theatre', 'by vibes', 'panic-cycle the VM') to lean declarative rules so the body better respects token budget.

Move the 20-item Danger Zones and detailed incident/reboot-heal histories into references/operations.md or a dedicated reference file, leaving the SKILL.md body as a navigable overview.

Add explicit validate-then-proceed checkpoints (and error-recovery loops) to the flat Deploy Commands and Agent Runner lifecycle sections to lift workflow_clarity to a 5.

Dimension	Reasoning	Score
Conciseness	The 534-line body is dense and mostly high-signal operational knowledge Claude would not know, but narrative flavor ('not healer theatre', 'by vibes', 'panic-cycle the VM') and lengthy incident rationales could be tightened, fitting 'mostly efficient but includes some unnecessary explanation' at 3 rather than the lean anchor at 5.	3 / 5
Actionability	Fully executable, copy-paste-ready commands and complete manifests throughout (Quick Health Check curl/kubectl one-liners, the full Redis AOF recovery Pod YAML, TypeScript job-spec generation, deploy scripts), matching the fully-executable anchor at 5.	5 / 5
Workflow Clarity	Key recovery workflows have explicit validation and feedback loops (Redis AOF: 'Wait, check logs, then clean up'; ADR-0244 verification window; rollout status in deploys), but several sections (flat Deploy Commands list, Agent Runner lifecycle) lack explicit error-recovery checkpoints, fitting 'clear sequence with most checkpoints present; minor validation gaps' at 4.	4 / 5
Progressive Disclosure	Well-sectioned body with a clearly signaled one-level-deep reference ([references/operations.md] linked with explicit context three times), but a large volume of inlined detail (20 Danger Zones, incident histories, Restate/Firecracker notes) that could be split into the single available reference file fits 'good structure; most content appropriately placed; minor organization gaps' at 4.	4 / 5
	Total	16 / 20 Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly states capabilities and provides an exhaustive trigger list, leaving no ambiguity about when to invoke it. The only soft spot is a few generic k8s trigger terms that could overlap with a broader Kubernetes skill.

Dimension	Reasoning	Score
Specificity	Lists eight concrete actions ('Deploy services, check health, debug pods, recover from restarts, add ports, manage Helm releases, inspect logs, fix networking'), matching the comprehensive multi-action anchor rather than the 1-2 action anchor at 3.	5 / 5
Completeness	Explicitly answers 'what' (operate the joelclaw cluster + enumerated actions) and 'when' via a dedicated 'Triggers on:' clause with concrete trigger phrases, satisfying the both-what-and-when anchor at 5.	5 / 5
Trigger Term Quality	Comprehensive natural-language trigger list including synonyms and tool names ('kubectl', 'pods', 'helm install', 'talosctl', 'colima', 'flannel', 'nodeport', 'PVC') plus service-specific phrases ('deploy redis', 'livekit helm'), matching the full-coverage anchor at 5.	5 / 5
Distinctiveness Conflict Risk	Tied to a named cluster with distinct tools (Talos/Colima/flannel), but several generic k8s triggers ('kubectl', 'pods', 'PVC', 'storage') carry minor overlap risk with a generic Kubernetes skill, fitting 'mostly distinct; minor overlap risk' at 4 rather than the minimal-risk anchor at 5.	4 / 5
	Total	19 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (535 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: joelhooks/joelclaw
Path: skills/k8s/SKILL.md
Commit: d952451

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.