Agent-retrieved covariates need three checks: date filtering alone lets encoder weights carry future facts
Language-model agents can retrieve fast-moving local context that survey baselines miss, but whether such covariates are usable for causal analysis needs separate verification. A pilot study anchored on Uganda's Youth Opportunities Program compares two retrieval workflows across 11 datasets and 47 treatment-outcome contrasts, with checks on retrieval timing, out-of-sample ranking and pretrained encoder weights. Neither workflow dominates, and swapping the encoder alone shifts downstream heterogeneity rankings.