Calibrating Generated Items Before Pretesting with Predicted Priors R package coldstart version 0.1.0 Predicts Rasch item difficulty from item features (text embeddings from any model, template family, content metadata) by ridge regression (Hoerl and Kennard, 1970, <doi:10.1080/00401706.1970.10488634>) with honest out-of-sample uncertainty, uses the predictions as robust informative priors to calibrate new items from small pretest samples, in the spirit of using collateral information in calibration (Mislevy, Sheehan and Wingersky, 1993, <doi:10.1111/j.1745-3984.1993.tb00422.x>), plans the pretest sample size needed for a target precision, and diagnoses item families whose predictions cannot be trusted. Software Depends: R (>= 4.1) Imports: stats Suggests: knitr, markdown Daniel Edi <danieledi2026@gmail.com> Comprehensive R Archive Network (CRAN) Daniel Edi [aut, cre, cph] (ORCID: <https://orcid.org/0000-0001-5475-819X>) MIT + file LICENSE (https://CRAN.R-project.org/package=coldstart/LICENSE) 2026-10-07 application/tgz https://CRAN.R-project.org/package=coldstart doi:10.32614/CRAN.package.coldstart