AI-based PHP vulnerability detection: a PRISMA 2020 systematic review — supplementary materials

Supplementary materials for a PRISMA 2020 systematic review of empirical studies (2018–2026) on PHP vulnerability detection using machine learning, deep learning, and large language models. Searches covered Scopus and OpenAlex; 58 studies (60 reports) met the eligibility criteria. This deposit includes the frozen protocol, search logs, screening and eligibility registers, the extraction/coding matrix, quality appraisal (five exploratory domains), the blind second-rater audit and its adjudication, VOSviewer bibliometric inputs, analysis scripts that reproduce every reported count, and a SHA-256 manifest. Anonymised for double-blind review; the non-anonymised version will replace it after review.