Skills Audit Report

Skills Audit Report

**Date:** 2026-02-15 **Auditor:** Automated Skill Quality Audit **Scope:** Recently added skills in business-growth/, finance/, marketing-skill/campaign-analytics/, project-management/

---

Executive Summary

The recently added skills fall into two distinct tiers:

1. **Business-growth, Finance, and Campaign Analytics skills** — Genuinely impressive. Production-ready Python tooling, deep domain frameworks, real structured outputs. These would make a domain practitioner say "this actually knows what it's doing."

2. **Project Management skills** — A mixed bag. The Atlassian-specific skills (jira-expert, confluence-expert, atlassian-admin, atlassian-templates) have strong knowledge-base content. The scrum-master and senior-pm skills are thin and generic. None of the PM skills have scripts or assets — they're pure prompt-engineering skills, which is a fundamentally different (and weaker) category.

**Overall: 4 POWERFUL, 1 SOLID, 4 SOLID, 2 GENERIC, 1 WEAK**

---

Detailed Skill Audits

---

1. business-growth/customer-success-manager

**Code Quality: EXCELLENT**

• 3 Python scripts (438 + 487 + 414 = 1,339 lines total)
• Well-structured: proper typing, argparse CLI, JSON/text dual output, error handling
• Zero external dependencies (stdlib only) — deliberate, documented design choice
• health_score_calculator.py: Multi-dimensional weighted scoring with segment-aware benchmarks (Enterprise/Mid-Market/SMB). Not placeholder math — real configurable thresholds, normalization logic, trend analysis
• churn_risk_analyzer.py: Behavioral signal detection with renewal urgency multipliers
• expansion_opportunity_scorer.py: Whitespace mapping and effort-vs-impact prioritization
• All scripts actually runnable with provided sample data

**Problem-Solving Quality: EXCELLENT**

• Health scoring framework reference (80+ lines) explains why each dimension is weighted as it is — genuinely pedagogical
• Real CS playbooks: not "be proactive" platitudes but specific intervention triggers (e.g., "if health score drops below yellow for 2 consecutive periods, escalate")
• QBR template is production-ready — has ROI summary tables, value-delivered sections, next-quarter planning
• Success plan template, onboarding checklist, executive business review — all structured with fill-in tables
• Uses real industry frameworks: DAU/MAU ratio, NPS scoring methodology, multi-threading depth

**Structure: STRONG**

• SKILL.md has proper frontmatter, TOC, input/output schemas, limitations section
• References are actually used by the scripts (health-scoring-framework.md maps directly to score calculation logic)
• Assets include sample data AND expected output JSON for validation

**Verdict: POWERFUL** ⭐ *Evidence: A CS leader could hand this to a team and they'd have a working health scoring system same day. The weighted scoring model with segment-aware thresholds is exactly how real CS platforms (Gainsight, Totango) work. The scripts produce struc