fix(skill): improve product-manager-toolkit per benchmark feedback (#54) (#102)
Addresses feedback from AI Agent Skills Benchmark (80/100 → target 88+):
SKILL.md restructured:
- Added table of contents for Progressive Disclosure Architecture
- Fixed second-person voice ("your" → imperative form throughout)
- Added concrete input/output examples for RICE and interview tools
- Added validation steps to all 3 workflows (prioritization, discovery, PRD)
- Removed duplicate RICE framework definition
- Reduced content by moving frameworks to reference file
New: references/frameworks.md (~560 lines)
Comprehensive framework reference including:
- Prioritization: RICE (detailed), Value/Effort Matrix, MoSCoW, ICE, Kano
- Discovery: Customer Interview Guide, Hypothesis Template, Opportunity
Solution Tree, Jobs to Be Done
- Metrics: North Star, HEART Framework, Funnel Analysis, Feature Success
- Strategic: Product Vision Template, Competitive Analysis, GTM Checklist
Changes target +8 points per benchmark quick wins:
- TOC added (+2 PDA)
- Frameworks moved to reference (+3 PDA)
- Input/output examples added (+1 Utility)
- Second-person voice fixed (+1 Writing Style)
- Duplicate content consolidated (+1 PDA)
Resolves #54
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>