zotero-slr-dedup-archived

A retired dedup workflow placed duplicate items into per-source 04-Superseded collections instead of merging them, keeping both records for PRISMA accountability. That workflow was retired on 2026-07-16 in favour of merge-with-type-normalization, leaving 71 item-memberships behind.

Use the zotero-merge-prep skill, which unions metadata and normalizes item types so Zotero's native Duplicate Items merge loses nothing. Run merge_prep.py scan on any collection key to find groups, then merge in the Zotero client.

Five of eleven sources spell the collection 04-Superceded, and the old script matched names exactly, so 15 items across wos, cao, and ssrn are invisible to a single-spelling query. Count both spellings; do not rename the collections.

No. Zotero's merge unions collection memberships, so the surviving record keeps every import collection and per-source counts stay invariant. The library already demonstrates this with a single record carrying acm, arxiv, and wos source memberships.

No. All executable scripts were deleted at retirement, including apply_dedup.py whose writes contradict the merge philosophy. This directory is documentation only; use zotero-merge-prep for any actual duplicate resolution.