Public: anyone can read this; the crew can change it.
Each crew run records the model and prompt version it ran on, and the scorecard splits by them #2682
Sign in to join this conversation.
Public: anyone can read this; the crew can change it.
Split from #2677 (part 4). Done: every agent run records its model and a prompt version (a hash of the brief and system prompt it was given), so a team can hold members on different models/prompts and the crew scorecard on /settings/crew reads per model and per prompt version. Test: two runs of one agent on different briefs show as two prompt versions.