Feedback on your pattern skill

[RichardHightower]

I took a look at your pattern skill and wanted to share some thoughts.

Links:

• GitHub
• Your skill in the SkillzWave marketplace

TL;DR

You're at 75/100, which lands you solidly in C territory. The skill has some really strong fundamentals—your Progressive Disclosure Architecture is genuinely good (28/30), and the way you've structured references with separate files for workflow, output templates, and verification is smart. But here's the thing: your frontmatter is broken, which tanks your Spec Compliance score to 0/15. That's a huge swing waiting to happen.

What's Working Well

• Progressive Disclosure architecture is chef's kiss. You've got SKILL.md as a tight overview, with four reference files at exactly one level deep (workflow.md, output-template.md, verification-guide.md, reference.md). That's the pattern to follow.
• Clear navigation signals throughout. TOCs where they belong, section headers with dashes, anchor links that actually point somewhere. Makes it easy to jump around.
• Strong verification structure. V1-V4 steps with confidence levels (high/medium/low) guide users on what to do next. That feedback loop is legit.
• Good token economy. SKILL.md stays focused—no bloat, each section earns its place. That's the kind of constraint that actually helps people.

The Big One: Fix Your Frontmatter

Your SKILL.md frontmatter is incomplete and has invalid YAML syntax. It cuts off mid-line and that's tanking your Spec Compliance score (0/15 right now). This is the quickest win you've got.

Current state (broken):

---
name: pattern
description: Learn design patterns from the current codebase
model: sonnet
allowed-tools: Bash, Glob, Grep, Read, Write
argument-hint: [pattern type, e.g., "api endpoint", "backgroun

What it should be:

---
name: pattern
description: Analyze and extract implementation patterns from codebases. Use when asked to "api endpoint pattern", "service pattern", "repository pattern", "background job", "authentication flow", or "how does this codebase implement X".
---

That one fix gets you +9 points right there—and it's literally just completing the YAML properly and adding concrete trigger phrases to your description. The description needs those specific terms so people actually discover the skill when they're asking about patterns.

Other Things Worth Fixing

1. Add an inline example to SKILL.md. Your Output Format section shows the template structure, but users have to jump to output-template.md to see a real, filled-in example. Drop a 3-5 line mini-example right in SKILL.md before linking them to the full template. (+1 point)

2. Trim the verification examples. Five detailed examples with full bash commands and results takes ~100 lines when 2-3 would do the job. Pick one passing case and one failure-recovery case; collapse the rest. (+1 point)

3. Flesh out trigger terminology in metadata. You've got good patterns listed in the supported patterns section, but the skill description itself is generic ("Learn design patterns from the current codebase"). Make it more specific in the frontmatter where people discover you first.

Quick Wins

• Fix frontmatter YAML → +9 points (spec compliance fix)
• Add trigger phrases to description → improves discoverability
• One inline example in SKILL.md → +1 point (utility)
• Tighten verification examples → +1 point (writing style)

You're close. Fix the frontmatter and add those trigger terms, and you're looking at pushing toward 85+.

---

Checkout your skill here: [SkillzWave.ai](https://skillzwave.ai) | [SpillWave](https://spillwave.com) 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.