cloud-deploy-blueprint

I've been digging through cloud deployment patterns, and your skill's approach to Blueprint-style infrastructure abstractions is exactly the kind of pragmatic thinking that keeps deployments from becoming maintenance nightmares—scoring 97 points makes sense given how well you've balanced the conceptual depth with actual usability.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 97/100, solid A territory. This is graded against Anthropic's Claude Skills best practices. Your Utility pillar maxed out at 20/20—those battle-tested patterns from your 9-hour deployment session are gold—while Progressive Disclosure Architecture at 28/30 shows excellent token economy and layering. The gap is in Spec Compliance (11/15), mostly around metadata discoverability.

What's Working Well

• Utility is locked in: You've captured real pain points (build-time vs runtime variables, Next.js env var patterns, managed service gotchas). The grep-friendly structure and copy-paste checklists make this immediately actionable.
• Reference architecture is chef's kiss: Three separate reference files at exactly one level deep keeps SKILL.md focused while giving people the full depth when they need it—that's solid progressive disclosure in practice.
• Battle-tested patterns shine: The GitHub Actions CI/CD pattern, pre-deployment checklist, and common gotchas section read like they came from actual war stories. That credibility matters.
• Metadata clarity: Your triggers (AKS/GKE/DOKS, CI/CD pipelines, Next.js vars) are specific enough that people searching for "managed services deployment" will actually find this.

The Big One: Add Trigger Phrases to Description

Right now your description is missing trigger phrases, which costs you 2 points in Spec Compliance and hurts discoverability. Your current description reads like a feature list:

End-to-end cloud deployment skill for Kubernetes (AKS/GKE/DOKS) with CI/CD...

It should lead with what the skill does when invoked:

Performs cloud deploy blueprint operations. Use when asked to "cloud deploy blueprint", "run cloud deploy blueprint", or "cloud deploy blueprint help". Covers end-to-end Kubernetes deployment patterns for AKS/GKE/DOKS with managed services and CI/CD integration.

This is low friction to add—just reword that frontmatter description—and it's a straight +2 points.

Other Things Worth Fixing

1. Add a TOC to SKILL.md (369 lines is long enough to benefit from one). After the Overview section, add anchored links to your major sections (Architecture Pattern, GitHub Actions CI/CD, Common Gotchas, Pre-Deployment Checklist). That's +1 point and makes navigation nicer.

2. Kill the second-person voice in a few spots. Line 220 has "check your cluster's node architecture"—flip it to "check the cluster's node architecture" for consistency with imperative form. Minor (+1 point), but it cleans up the voice.

3. Consider a brief intro paragraph before diving into "When to Use This Skill." Just 2-3 sentences on what this skill solves (not a feature list, but the problem it addresses) would help context for people skimming.

Quick Wins

• Fix description → add trigger phrases (+2 points)
• Add TOC to SKILL.md (+1 point)
• Adjust second-person voice (+1 point)
• These three changes push you to 101/100 (with that +5 modifier, you're already sitting pretty)

Checkout your skill here: SkillzWave.ai | SpillWave We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.