Decision Space Expansion — Eval Report Hidden ground truth: hand-authored gold option sets per scenario (never revealed to the systems under test). Scenario Method Options found Gold hit Coverage Blind spots missed A single_model 1 1 20% 4 A naive_multiagent 3 3 60% 2 A decision_workbench 5 5 100% 0 B single_model 1 1 25% 3 B naive_multiagent 3 3 75% 1 B decision_workbench 4 4 100% 0 C single_model 1 1 33% 2 C naive_multiagent 3 3 100% 0 C decision_workbench 3 3 100% 0 Aggregates Method Total gold hit (of 12) Avg coverage single_model 3/12 26% naive_multiagent 9/12 78% decision_workbench 12/12 100% Blind spots recovered by Decision Workbench Scenario A: crm-as-a-feature, spreadsheet-plus-automation Scenario B: membership-platform Scenario C: (none beyond baselines)
