Hey people, I’m working on an early OSS tool called vero-eval, for testing and debugging RAG/Agents Since we’re early, I’d love to hear what actually matters to you all in an eval tool: something useful for you specifically or that one feature you wish others had. Repo (pip install available) is here if you want to poke around: https://github.com/vero-labs-ai/vero-eval Would appreciate any feedback. Thanks already