A Verbatim-Grounded, Field-Neutral Tag Layer for Cross-Disciplinary Reading of Academic Literature

A field-neutral, verbatim-grounded tag layer for academic literature. Every paper is described with the same small set of faceted tags (claim type, phenomenon as a normalised cross-field join key, claim relations, replication status, limitations); the design rule is that every tag is licensed by an exact sentence from the paper's own text, and the audit reports each tag's provenance status rather than assuming full exact-body grounding. A proof-of-format on a curated 126-paper corpus spanning 39 disciplines (116 fully faceted). Version 8.0 added a systematic audit of all 519 statistical-tag assignments (420 accepted, 99 rejected, 19.1% error rate) with the statistics section recomputed, a corrected complexity analysis backed by a real inverted index, a JSON Schema, a reproducible build pipeline, and regression tests. Version 9.0 adds a per-paper redistribution-rights manifest (each source's licence verified against its landing page and the Crossref API): full text is reproduced only for the 19 papers under a verified redistributable licence (CC-BY / CC0 / public domain); every other paper shows the tag layer with short licensing excerpts and links to the source, not the reproduced text. It also states the audit method (rule-based with LLM-assisted adjudication), notes the absence of an inter-rater reliability study as a limitation, and reconciles all release metadata.