visualization-patterns

I took a look at your visualization-patterns skill and noticed you're tackling a pretty common pain point—helping developers translate data into clear, visual representations without getting lost in implementation details. The structure's solid (71/100), though there's some room to tighten up how you're guiding users through the progression of complexity.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 71/100, C grade—solid fundamentals but needs some depth work. This evaluation is based on Anthropic's best practices for skill design. Your strongest area is Spec Compliance (12/15)—clean YAML frontmatter, proper naming conventions. Weakest area is Utility (12/20)—you've got the framework sketched out, but it needs concrete examples and validation loops to actually help someone build a dashboard.

What's Working Well

• Spec compliance is tight – Valid YAML, proper hyphen-case naming, all required fields in place
• Clear trigger phrases – "dashboard", "reports", "narratives" are good GTM-specific keywords that'll help discoverability
• Consistent terminology – KPI, dashboard, visualization language stays consistent throughout, which matters when users are trying to follow along
• Audience-first framing – Starting with "Audience & Story" is smart; you're not jumping straight into technical implementation

The Big One: Missing Reference Files and Templates

This is what's holding you back most. Your skill lists templates ("Dashboard wireframe grid with KPI slots", "Metric dictionary", "Adoption checklist") but doesn't actually provide them. Right now, a Claude agent reading this has to invent those templates from scratch, which defeats the whole purpose of a skill.

The fix: Create three reference files:

• references/dashboard-wireframe.md – Actual template with grid layout, KPI slot descriptions, and real examples
• references/metric-dictionary-template.md – YAML or table format showing definition, source, owner, refresh schedule with 2-3 examples
• references/adoption-checklist.md – Actual checklist items (not just the concept of a checklist)

This alone gets you +5 points and transforms the skill from guidance to genuinely actionable.

Other Things Worth Fixing

1. Workflow steps need sub-steps – "Audience & Story" is one line. Break it into: identify persona, document decision cadence, list 3-5 key questions, define success metrics. This makes it executable rather than just conceptual. (+3 points)

2. Add a validation section – After step 5, there's nothing on testing with actual users, verifying data accuracy, or iterating. "Does it work?" is a pretty important question. (+2 points)

3. Chart selection needs examples – "Match metric type to chart" is vague. Create a quick reference: Trend=Line chart, Composition=Stacked bar, Comparison=Grouped bar. Real examples, not abstract principles. (+2 points)

4. Beef up the description – Only 1-2 trigger phrases. Add: "chart selection", "KPI design", "analytics dashboards", "data storytelling" to catch more searches.

Quick Wins

• Create three reference files with actual templates (+5 points)
• Add numbered sub-steps under each framework stage (+3 points)
• Add a validation/feedback loop section (+2 points)
• Expand description with more trigger phrases (+1-2 points)

These four changes get you from 71 to ~84 territory—solid B-grade work.

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.