← ClaudeAtlas

textual-criticismlisted

Explain why translations differ on whether a verse belongs in the Bible, and what the manuscripts actually say. Use when the user asks about a footnote or a missing verse, e.g. "why does my Bible say some manuscripts omit this", "is Mark 16:9-20 original", "the woman caught in adultery John 8", "1 John 5:7 Comma Johanneum", "what do the earliest manuscripts say".
kalib-code/openlumin-skills · ★ 1 · DevOps & Infrastructure · score 69
Install: claude install-skill kalib-code/openlumin-skills
# Textual criticism Work under `bible-evidence-grounding`. The person asking this question is very often anxious rather than academic — they found a footnote and want to know whether their Bible can be trusted. Tone is the substance of this skill. ## Know what the tools can and cannot give you `get_textual_evidence` takes a `reference` and the user's `question` and returns evidence cards. Verified against production, here is what actually comes back: - **Original-language card** — interlinear tokens for the passage from OpenLumin's own STEPBible-derived tables. - **Ancient witness card** — rows from the stored versional corpora, named by edition: ETCBC/BHSA [Leningrad] for the Masoretic Text, ETCBC/Peshitta, Samaritan Pentateuch [Chester Beatty 751], Sefaria Targum [Onkelos, Jonathan], Clementine Vulgate, and MorphGNT / SBLGNT for the Greek New Testament. - **Source availability card** — registry metadata about which corpora are licensed and loaded. **It does not name manuscript sigla, give manuscript dates, or state which manuscripts include or omit a reading.** Verified: Deuteronomy 32:8 returns five source groups, Isaiah 7:14 four, Psalm 22:16 two — and in each case the tool explicitly declines to say whether they agree, because "different scripts/languages … requires alignment review". For the **New Testament** only one source group exists (SBLGNT), and the tool says so: "only one source group is available, so no cross-witness alignment claim is made." John