C1: An Expert-Annotated Dataset of Security-Relevant Product Backlog Items

288 product backlog items independently labeled for security relevance by nine security practitioners at a large industrial enterprise, yielding 972 binary assessments with substantial inter-rater reliability (Fleiss' κ = 0.787, raw agreement P₀ = 0.800). 80 items (27.8%) are security-relevant; security-relevant items additionally carry optional Firesmith security-requirement category metadata. Items are distributed as JSONL with aggregated majority-vote labels. The item texts derive verbatim from the TAWOS dataset (Tawosi et al., MSR 2022; Apache License 2.0) and originate from the public Atlassian JRASERVER and CONFSERVER issue trackers, filtered through a deduplication pipeline described in the associated paper. This dataset is contribution C1 of: Ignacio García N., Florian Angermeir, Fabiola Moyón C. "From Backlog Items to Security Guidance: Towards Continuous Security Compliance." 41st IEEE/ACM International Conference on Automated Software Engineering (ASE 2026), Industry Showcase. See the bundled README for schema and usage examples.