community-sentiment-dashboard

I found your community-sentiment-dashboard skill and noticed it's tackling a really useful problem—real-time sentiment tracking across communities—but the implementation feels like it's caught between a few different architectural directions. With a 62/100, there's solid potential here, but I'd like to dig into what's holding it back.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 62/100, which lands you in D territory. This is based on Anthropic's five-pillar framework for agentic skills. Your Spec Compliance is actually solid (11/15), but Utility is dragging things down (11/20)—you've got the structure, just missing the substance.

What's Working Well

• Valid frontmatter – Your YAML is clean and follows conventions. Hyphen-case naming is correct.
• Clear problem domain – The five-component framework (Signal Sources, Sentiment Stack, Risk Board, Advocates, Actions) is a sensible way to slice sentiment tracking.
• grep-friendly structure – The skill file is organized in a way that's easy to search and navigate.

The Big One: Missing Templates and Examples

Here's the thing: you mention templates but they don't actually exist. Your Templates section lists "Dashboard slide layout (KPI tiles + trend graphs + highlights/lowlights)" and "Escalation tracker" and "Advocate tracker," but users get nothing concrete to work with.

Why it matters: Without actual examples, users can't visualize what the output looks like or understand the exact shape of data they're building. That's a quick +8 point swing right there (hitting PDA layering, reference depth, and utility all at once).

The fix: Create reference files—references/dashboard-template.md with an actual slide layout showing sample KPIs, references/escalation-tracker.md with a real spreadsheet structure, references/example-dashboard.md with sample sentiment data. Don't overthink it; even a markdown table showing "5,234 mentions | 68% positive | 12 escalations" moves the needle.

Other Things Worth Fixing

1. Trigger phrases are buried – Your description doesn't say "Use when asked to 'sentiment dashboard' or 'community listening report'." Users won't find this skill unless you're explicit. Rewrite the description to include actual trigger terms users would search for. (+2-3 points)

2. Workflow steps are missing – You list components but no sequential workflow. Add numbered steps: 1. Configure signal sources (connect APIs), 2. Set up sentiment analysis stack, 3. Build risk board, 4. Identify advocates. (+3-5 points)

3. Imperative voice inconsistency – Sections use noun phrases ("Signal Sources – social listening feeds") instead of verbs ("Configure Signal Sources – Connect social listening feeds, integrate community platforms"). Tighten the language. (+2 points)

Quick Wins

• Add reference files with actual templates and example data (+8 points—biggest bang for your buck)
• Rewrite the frontmatter description with explicit trigger phrases (+3 points)
• Convert framework to numbered workflow steps with concrete actions (+4 points)
• Show a sample dashboard output in the examples section (+3 points)

These four changes alone could push you to mid-70s.

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.