kmjy98-sketch
Community@kmjy98-sketch
Korean legal study and contract-review skill suite covering law exam prep, Anki cards, PDF ingestion, and playbook-based NDA/contract analysis.
Agent Skills by kmjy98-sketch
Showing 25 vetted skills indexed across 1 GitHub repositories.
글다듬기
Reviews and refines Korean prose against nine grammar, style, and structure principles.
transcript-tools
Splits long Markdown transcripts into evenly sized part files using a Python script.
spaced-repetition
Schedules review items and tracks recall scores using the SM-2 spaced repetition algorithm.
legal-response
Generates legal response drafts from review reports using playbook-based posture templates.
textbook-problem-intake
Scans textbook PDF extracts for embedded case and multiple-choice problems and registers them into a problem index.
whisper-transcribe
Transcribe Korean audio lectures to text using Whisper on Colab GPU or local CPU.
socratic-loader
Extract textbook pages, transcripts, and related problems for Socratic study sessions.
pdf-ingest
Extract text and tables from native and scanned PDFs into markdown chunks.
socratic-core
Runs Socratic tutoring sessions for Korean law study with scaffolding, hints, and grading.
progress-tracker
Tracks study progress by managing part completion states in a progress.json file.
anki-card-generation
Converts Korean law study materials into Anki Basic and Cloze flashcards with source verification.
study-notes
Generates structured Korean law study notes from transcripts, textbooks, and past exam questions.
legal-briefing
Converts contract review reports into meeting-ready legal briefing memos.
problem-index
Manage a Korean law problem index with lecture-based queries, PDF extraction, and issue frequency analysis.
legal-review
Review contracts and NDAs against a playbook, classifying clauses as GREEN, YELLOW, or RED.
transcript-correction
Corrects Korean lecture transcripts against textbook sources and tracks correction state in a JSON log.
file-classification
Classify and organize Korean law study files into subject folders using keyword routing rules.
case-answer-review
Grades Korean law case-study answers and generates review packets from textbook indexes and RAG results.
file_ops_log
Logs file moves, renames, copies, and trash operations with paths and SHA-256 hashes.
pdf-export
Convert Markdown study notes into A4 PDF documents using Playwright and Chromium.
auto-index
Extract statutes, cases, and keywords from Markdown notes into a tag index.
legal-compliance
Generates playbook-based compliance checklists and required actions from contracts and legal review reports.
textbook-reflow
Restructures OCR-extracted Korean law textbook markdown into Obsidian-optimized layouts.
law-note-supplement
Augments, reviews, and restructures existing Korean law study notes with sourced citations.
Frequently Asked Questions About kmjy98-sketch
FAQPage SchemaWhat tasks can I accomplish with kmjy98-sketch's skills?▼
You can review contracts and NDAs with clause-level risk ratings, generate legal briefings and compliance checklists, transcribe lectures via Whisper STT, convert PDFs and scanned textbooks to markdown, build Anki cards, track study progress, and draft Korean legal documents such as case-record answers and briefs.
Who are these skills designed for?▼
They target Korean law students preparing for bar-style exams (사례형, 기록형), legal professionals handling contract review and compliance checks, and self-directed learners who need spaced-repetition scheduling, progress tracking, and structured study notes from textbooks and lecture transcripts.
How do the legal review skills work in practice?▼
The legal-review skill reads PDF, DOCX, TXT, or MD documents and evaluates each clause against a playbook, marking status GREEN, YELLOW, or RED. Downstream skills then produce response drafts, executive briefing memos, or privacy and data-retention compliance checklists from that report.
What are the prerequisites and dependencies?▼
PDF ingestion relies on pypdf, pdfplumber, and easyocr for scanned documents; markdown-to-PDF export uses Playwright with Chromium rendering and Korean fonts; Whisper transcription recommends a Colab GPU runtime. Study skills expect an Obsidian-style vault with progress.json and problem_index.json files.
Are these skills free and open source?▼
The manifest lists 26 publicly published skills under the kmjy98-sketch account with no stated licensing fees or paid tiers. Each skill is distributed as a folder with native frontmatter metadata, so users can inspect descriptions and triggers before adopting them locally.