The Validator skill serves as the final synthesis layer for Claude Code, transforming raw analytical data into an unambiguous, professional report. It evaluates technical proposals, code changes, or architectural designs by assigning a clear verdict—GOOD, NEEDS MAJOR WORK, or BAD—and documenting specific strengths, critical flaws, and hidden blindspots. By reframing narrow problem definitions and providing measurable 'bulletproof' success criteria, this skill ensures that teams move forward with high confidence and a concrete roadmap for improvement.
Key Features
01Detailed 'Path Forward' guidance tailored to specific report outcomes
02Critical flaw identification with business and technical impact analysis
03Unambiguous verdict system (GOOD/NEEDS MAJOR WORK/BAD) with confidence levels
04Measurable success criteria checklists for bulletproof implementation
051 GitHub stars
06Assumption-challenging section to surface hidden blindspots and dependencies