@wundr.io/agent-eval

Agent evaluation framework with LLM-based grading for AI agent quality assessment