AI-Run Modern LaTeX Manuscript Workflow and Replication Packet

Workflow and replication packet for the Modern LaTeX Editions of Mathematics Manuscripts project. This record documents the practical AI-run pipeline used to route public-domain scans and existing TeX/OCR into modern LaTeX working editions, translations, audit packets, and Zenodo/GitHub releases. Latest update. Adds workflow_addendum_20260612_source_identity_repair_worklists.md, a public-safe methodology note distilled from the latest Weber, SGA, Noether, and Kneser repair lanes. It records edition/source identity checks, structured repair-worklist design, OCR witness discipline, source-derived figure policy, language propagation after source closure, and public archive labelling rules. Important distinction. The project is AI-run, not merely AI-assisted, but public releases distinguish source-checked drafts/translations from OCR/crop/formula witness packages and raw provenance. OCR and formula recognition are treated as locator layers rather than textual authority. Corrections and collaboration. Public working mirror and issue/pull-request surface: https://github.com/KokunoYumeto/modern-latex-manuscripts.《数学手稿现代LaTeX版》项目的工作流与复现数据包。本记录详述了实用化的AI驱动工作流，该工作流用于将公有领域扫描件与现有TeX、光学字符识别（Optical Character Recognition，下文简称OCR）文件处理为现代LaTeX可编辑工作版本、译稿、审计数据包及Zenodo与GitHub发布包。 最新更新。本次更新新增了workflow_addendum_20260612_source_identity_repair_worklists.md文件，该文件是从最新的Weber、SGA、Noether及Kneser修复流程中提炼出的面向公开场景的方法论说明文档，涵盖版本/来源身份核验、结构化修复工作清单设计、OCR佐证规范、来源衍生插图策略、源文件定稿后的语言传播规则及公开档案标注规范。 重要区分说明：本项目采用AI全流程驱动，而非仅AI辅助；但公开发布内容会区分经来源核验的草稿/译稿、OCR/裁切/公式佐证数据包及原始来源数据。 OCR与公式识别仅被视为定位图层，而非文本权威依据。 更正与协作渠道：公开工作镜像及问题/拉取请求入口：https://github.com/KokunoYumeto/modern-latex-manuscripts。