Skills Directory

This skill should be used when the user wants to "run an evaluation", "evaluate my ADK agent", "write an eval dataset", "analyze eval failur

Implement comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking. Use when testi

Formal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles

Patterns and techniques for evaluating and improving AI agent outputs. Use this skill when: - Implementing self-critique and reflection loop

Blocks a production model change until a private eval has been run. Use before any model version bump or provider switch.