Use this when you need to EVALUATE OR IMPROVE or OPTIMIZE an existing LLM agent's output quality - including improving tool selection accuracy, answer quality, reducing costs, or fixing issues where the agent gives wrong/incomplete responses. Evaluates agents systematically using MLflow evaluation with datasets, scorers, and tracing.
Install
npx skills add mlflow/skills --skill agent-evaluationnpx skills add mlflow/skills --skill agent-evaluation -a claude-code -gnpx skills add mlflow/skills --skill agent-evaluation -a codex -ggemini skills install https://github.com/mlflow/skills.git --path agent-evaluationgh skill install mlflow/skills agent-evaluationgit clone --depth 1 https://github.com/mlflow/skills /tmp/mlflow__skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/mlflow__skills/agent-evaluation ~/.claude/skills/agent-evaluation # Claude Code
cp -R /tmp/mlflow__skills/agent-evaluation ~/.agents/skills/agent-evaluation # Codex, Cursor, Gemini CLI, Copilot, OpenCode