What problem does it solve? Unsorted uploads pile up in an inbox folder, and manually sorting law school materials (textbooks, transcripts, exams) into the right subject and instructor folders is repetitive and error-prone. This Skill automates that classification with deterministic rules, dry-run previews, and lock-based protection against rework. ## Core Features & Use Cases - Domain-separated classification: Routes files into legal domains (민사/형사/공법/선택법) or the admission domain (로스쿨입시/LEET) based on filename keywords, with instructor-specific routing overrides. - Safe execution model: Defaults to DRY-RUN, excludes protected roots (shared drives, trash, RAG data, original archives), and supports a lock registry keyed by relative path plus mtime to skip completed files and flag modified ones for re-review. - Naming and rule references: Ships classification rules and naming conventions (고유명사 vs 일반명사, mandatory year suffixes, folder-based type management) as reference documents. - Use Case: After uploading a batch of lecture PDFs and transcripts to 5.기타/_inbox, run the v3 script in dry-run mode to preview where each file will move, then execute to relocate them into the correct subject and instructor folders. ## Quick Start Ask the AI to run the file-classification script in dry-run mode on the inbox folder for the legal domain, review the planned moves, and then execute the classification.
