Pre-Deployment Verification of LLM-Generated Robot Artifacts: A Systematic Review — Data and Code Archive

Data and code archive for the systematic review "Pre-Deployment Verification of LLM-Generated Robot Artifacts: A Systematic Review" (single author; manuscript under review). Version 4 updates the archive to the 77-study corpus reported in the manuscript. It adds the data layer of record (data/v4_corpus_77/corpus_master.csv) with the matching codings, per-study Q1–Q5 rule record and ceiling reasons; the citation search and venue re-check records (audit_trail/v4_citation_search/); the log of changes from version 3 (audit_trail/v4_revision/REVISION_LOG_v4.md); the 72-record held-out preprint frontier; and the scripts that generate and check every table (scripts/v4/). Files from version 3 are retained unchanged as the record of the 67-study stage. Contents:- Screening and extraction data, including PRISMA 2020 flow counts and the verified extraction layer.- Coding instrument: frozen coding rubric (guarantee decision procedure Q1–Q5, family definitions, integration test) and the per-study coding records.- Search documentation: versioned search strings for Scopus, Web of Science, IEEE Xplore, and arXiv (PRISMA-S), and the citation search.- Scripts: table generation, numeric consistency checks, reliability computation, and figure generation.- MANIFEST with SHA-256 hashes of all archived files. Licenses: data and documentation under CC BY 4.0; scripts under the MIT license (scripts/LICENSE).