The archive's predictions as data: 4,385 claims (forecasts, standing claims, unbuilt tests, historical theses and priority records) read out of every deposit and its attachments, each registered at its earliest statement, with every later statement quoted, its dated revisions, its series and the records that claim it came true or failed. The 194 subject pages that read it live at operativesemiotics.org/predictions. Generated by operative-semiotics scripts/index/build_index.py; never hand-edited. Earlier passes stand beside it: prediction-ledger (falsification conditions and resolved outcomes) and prediction-registry.json (the August extraction whose counts this supersedes).