test_that("stress testing is reproducible", { ex <- autogenai_example() s1 <- stress_test(ex$task, ex$data, ex$providers[[1]], ex$prompts[[1]], seed = 7) s2 <- stress_test(ex$task, ex$data, ex$providers[[1]], ex$prompts[[1]], seed = 7) expect_equal(s1$quality, s2$quality) expect_true(is.numeric(robustness_score(s1))) })