Summary

This PR adds a new agentic-eval skill to the skills collection, focused on patterns for evaluating and improving AI agent outputs.

Skill Contents

• Reflection Pattern: Self-critique and iterative improvement loops
• Evaluator-Optimizer Pattern: Separate generation/evaluation components
• Code-Specific Reflection: Test-driven refinement workflows
• Evaluation Strategies: Outcome-based, LLM-as-Judge, Rubric-based
• Best Practices: Clear criteria, iteration limits, convergence checks

Use Cases

• Implementing self-critique and reflection loops
• Building evaluator-optimizer pipelines for quality-critical generation
• Creating test-driven code refinement workflows
• Designing rubric-based or LLM-as-judge evaluation systems
• Measuring and improving agent response quality

This skill is domain-agnostic and can be applied to any AI agent system requiring output quality improvement.