Bayesian Two-Stage Design with Window-Cohort and Controlled Roll- on for Time-to-Event Estimand // Calibrates Bayesian two-stage designs for single-arm phase II trials with time-to-event endpoints using a window-cohort with controlled roll-on. Interim monitoring is anchored to a locked interim cohort and a pre-specified follow-up requirement, so analysis timing remains predictable while preserving follow-up maturity. The package searches feasible interim rules, optimizes final sample size and decision thresholds, evaluates operating characteristics by Monte Carlo simulation, and supports exponential, Weibull, log- normal, log-logistic, and user-defined baseline survival models. Related published foundations include Simon (1989) doi:10.1016/0197-2456(89)90015-9 and Cotterill and Whitehead (2015) doi:10.1002/sim.6426.