What problem does it solve? Preparing a Socratic law study session requires manually locating the right textbook pages, parsing lecture transcripts for page references, and finding related practice problems. This Skill automates that preparation by parsing transcripts, extracting and caching textbook PDF text, and searching a problem index and RAG store. ## Core Features & Use Cases - Transcript Page Parsing: Extracts (페이지 X) page references from lecture transcripts to determine the study page range. - Textbook PDF Extraction with Caching: Pulls the relevant page range from textbook PDFs using pypdf and caches results as Markdown for reuse. - Problem and RAG Search: Queries a problem index JSON and qmd-based BM25/vector search to surface related multiple-choice and case-type exam questions. - Batch Extraction: Chunks entire textbook PDFs into 30-page segments with legal keyword extraction (statutes, case citations) and builds a chunks index for RAG. - Use Case: A law student starts a study session on 동시이행항변권; the loader parses the transcript, extracts textbook pages 120-145, and returns all related practice problems for the session. ## Quick Start Run the socratic loader script with a transcript file, a textbook PDF path, and a cache directory to prepare my study session.
