What problem does it solve? OCR and PDF extraction pipelines produce messy markdown with broken sentences, page headers, and flattened hierarchies that render poorly in Obsidian. This Skill reflows that raw output into a clean, logically structured document without altering legal citations. ## Core Features & Use Cases - Noise Removal and Rejoining: Strips repeated page headers/footers, stray page comments, and OCR spacing errors, then rejoins sentences split across page breaks. - Legal Structure Visualization: Converts inline enumerations into markdown lists, rebuilds scholarly debates as comparison tables, and isolates statutes and precedents in Obsidian callouts like [!조문] and [!판례]. - Integrity Verification: Counts statute and case citation identifiers before and after reflow to guarantee nothing was dropped, and preserves YAML frontmatter byte-for-byte. - Use Case: After running pdf-ingest on a civil law textbook chapter, ask for a reflow to get a study-ready Obsidian note with heading hierarchy, precedent callouts, and theory comparison tables. ## Quick Start Reflow the attached OCR-extracted civil law textbook markdown into an Obsidian-optimized structure while preserving all case and statute citations.
