decision-analysis

I took a look at your decision-analysis skill and wanted to share some thoughts.

Links:

• GitHub
• Your skill in the SkillzWave marketplace

The TL;DR

You're at 74/100, which puts you in C territory (70-79 range). This is based on Anthropic's skill best practices. Your strongest area is Writing Style (10/10 — really clean, instructional voice), but Progressive Disclosure Architecture is dragging you down at 13/30. That's the biggest opportunity here.

What's Working Well

• Writing is solid — Imperative voice, no marketing fluff, good density. You're not over-explaining or being wishy-washy.
• Utility is maxed out (20/20) — Your templates, decision tables, and sensitivity analysis examples are genuinely useful. This skill solves real problems.
• Workflow clarity — The numbered steps and checklists make it straightforward for developers to know what they're doing. That's harder than it looks.
• Metadata is mostly there — Valid YAML, correct naming conventions, appropriate tool access.

The Big One: Progressive Disclosure Architecture (13/30)

Here's the problem: You've crammed everything into a single 572-line SKILL.md file. For a skill this substantial, you need a layered, reference-based structure.

Why it matters: Developers using your skill need to quickly find what they need. Long, single-file skills feel overwhelming and waste token budget for agents.

The fix:

1. Create a references/ directory in your skill folder
2. Move detailed templates (decision tables, weighted scoring matrices, decision tree examples) into separate .md files
3. Keep SKILL.md as the core interface — keep it under 300 lines
4. Link to references using clear section headers

Example structure:

decision-...