Evaluate code, decisions, or content across 36 dimensions with φ-bounded confidence.
npx skills add https://github.com/zeyxx/CYNIC-legacy --skill cynic-judge
Please help me install this Agent Skill. Skill: cynic-judge Source: https://github.com/zeyxx/CYNIC-legacy/tree/main/.crush/skills/cynic-judge Command: npx skills add https://github.com/zeyxx/CYNIC-legacy --skill cynic-judge
This Skill provides a rigorous, multi-dimensional framework for evaluating the quality of code, decisions, or content, ensuring a balanced and objective assessment rather than a superficial one.
Use the cynic-judge skill to evaluate the provided code snippet for quality.
High-intent search queries and answers about installing and using this skill.
Multi-dimensional code review evaluates items across 36 dimensions grouped into 5 core axioms to ensure balanced assessment. It calculates a geometric mean Q-Score so no single strong dimension compensates for a weak one.
A φ-bounded confidence rating guarantees that confidence in any quality assessment judgment never exceeds 61.8%. This enforces intellectual humility during objective evaluation by preventing absolute certainty in scoring outcomes.
You score business proposals by applying a 36-dimension evaluation framework using 5 core axioms. This generates a holistic geometric mean Q-Score that reflects overall viability without letting single strong metrics skew the result.
Yes, this framework applies to general decision making, content quality assessment, and critical review tasks. It requires adherence to 5 axioms and outputs a geometric mean Q-Score with maximum confidence capped at 61.8%.
The limitation of a geometric mean Q-Score in quality assessment is that no single strong dimension can compensate for a weak one. Scores are strictly bounded by a 61.8% maximum confidence to enforce objective evaluation.
No, the cynic-judge skill requires no external dependencies to perform quality assessments. It operates independently using its internal 36-dimension scoring system and 5 core axioms to generate objective evaluation results.
yangqiongCleans raw datasets by handling missing values, duplicates, and data types.
ovachieverEvaluate scholarly work using the ScholarEval framework with dimension-based scoring.
NabilThangeAutomate linting and static analysis for Node.js, Python, and TypeScript code.
JupiterTheWarlockGuide designers in removing unnecessary complexity from interfaces.
