crm-hygiene

I've analyzed the crm-hygiene skill at 65/100 and here's an opening that feels genuine:

Took a look at your CRM hygiene skill—solid attempt at tackling data quality, though it feels like you're doing heavy lifting in areas that might benefit from more upfront validation. The 65/100 reflects some real gaps in Progressive Disclosure architecture that could cut your token usage significantly.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 65/100, squarely in D territory. This is based on Anthropic's skill best practices rubric. Your strongest area is Spec Compliance (11/15)—the frontmatter is valid and naming conventions are solid. But you're bleeding points in Progressive Disclosure (18/30) and Ease of Use (16/25) because the skill lacks supporting reference files and trigger phrases that would actually get people to use it.

What's Working Well

• Valid frontmatter structure – Your YAML is clean and follows conventions properly
• Solid problem domain – CRM data quality is a real pain point; the framework approach (Stage Definitions → Validation → Monitoring) is logical
• Professional tone – No marketing fluff; stays instructional throughout
• Reasonable conceptual framework – The 5-step structure gives users something to hang their hat on

The Big One: Missing Reference Files

This is your biggest drag on score. You mention three templates—Stage checklist, Hygiene scorecard, Escalation request form—but they're just listed, not provided. That means all the detail has to live in the main skill file, which bloats it and wastes tokens on context that could be delegated.

Here's the fix: Create actual reference files:

• references/stage-checklist-template.md with fields, approvers, entry/exit criteria
• references/hygiene-scorecard.md with sample scoring
• references/validation-queries.md with actual SQL or query patterns

This alone gets you +5 points and makes the skill actually reusable. Right now it reads like a lecture; with templates it becomes a toolkit.

Other Things Worth Fixing

1. Add trigger phrases to the description – "audit CRM", "pipeline inspection", "data quality check" aren't in there. Your current description is vague ("enforce stage definitions"). Users won't know to invoke this. +2 points

2. No validation workflow – You mention monitoring but never show the actual run→check→fix pattern. Add a "Validation Steps" section with concrete checks (count missing next-steps, identify aging deals, re-run until <5% violations). +3 points

3. Vague tips section – "drive accountability" doesn't tell anyone what to do. Change "Share hygiene scores transparently to drive accountability" to "Publish weekly hygiene scores to #sales-ops Slack with top/bottom performers." +2 points

Quick Wins

• Add 3-4 trigger phrases to frontmatter description
• Create 2-3 reference template files in references/
• Replace conceptual tips with concrete actions
• Add a validation checklist before finalizing hygiene runs

These changes realistically get you to 80-85/100 with minimal effort.

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.