LLM evaluation framework for Ruby, powered by RubyLLM. Tribunal provides tools for evaluating and testing LLM outputs, detecting hallucinations, measuring response quality, and ensuring safety. Perfect for RAG systems, chatbots, and any LLM-powered application.