Conversational agent that guides you through prompt evaluation and iterative improvement with multi-metric scoring.
Not persisted. If you set env vars in .env.local, you don't need these.
I'll help you systematically evaluate and improve your prompts using automated experiments and multi-metric scoring.