About Rmasood1122/eval-conductor
The integrity layer for your evals. Your evals already produce numbers — from claude plugin eval, pytest, promptfoo, or your own scripts. This proves them: that the verdict is real, the bar was fixed before the run, and the decision is tamper-evident. It sits next to MLflow, LangSmith or Braintrust; it doesn't replace them. It answers the one question none of them do