call-analysis-framework

I found your call-analysis-framework and noticed you're tackling the pretty complex problem of analyzing sales calls systematically—that's the kind of thing that usually requires some careful architectural thinking. At 60/100, there's real potential here, but the execution could be tightened up in a few key areas to make it significantly more useful.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 60/100, solid D-grade territory. This is based on Anthropic's best practices for skill architecture. Your Spec Compliance (11/15) is actually pretty strong—the YAML is clean and structure is valid. But Utility (10/20) and Progressive Disclosure Architecture (15/30) are dragging the score down significantly. You've got conceptual bones here, but they need flesh on them.

What's Working Well

• Valid metadata structure – Your frontmatter is properly formatted with required fields; no technical debt there
• Clear framework concept – The five-step structure (Section Definition → Scoring Rubric → Signal Tags → Templates → Routing) shows solid thinking about call analysis
• Objective tone – No marketing fluff, purely instructional; you're focused on functionality, not hype
• Problem awareness – You identified a real need (call analysis for sales) that actually matters

The Big One: Missing Reference Files and Implementation Details

Here's what's killing your utility score: everything is 31 lines of conceptual description with zero implementation details. You list "Scoring Rubric – 1-5 scale" but don't show what that actually looks like. You mention "templates" but provide zero actual templates.

Concrete fix: Create a references/ directory structure:

• references/scoring-rubric.md – Actu...