langtutor-vocab

Manages vocabulary learning with spaced repetition, mnemonics, and personalized flashcard exercises.

Updated Jan 12, 2026
One-click install
npx skills add https://github.com/Calcifer777/learn-korean --skill langtutor-vocab-calcifer777
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: langtutor-vocab
Source: https://github.com/Calcifer777/learn-korean/tree/main/.claude/skills/langtutor-vocab
Command: npx skills add https://github.com/Calcifer777/learn-korean --skill langtutor-vocab-calcifer777

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Language learners struggle to retain vocabulary long-term because they study words in isolation, cram without spacing, and lack structured review schedules. This Skill implements evidence-based vocabulary acquisition using the SM-2 spaced repetition algorithm, multi-modal encoding, and production-focused exercises. ## Core Features & Use Cases - Spaced Repetition Reviews: Tracks each word's interval, ease factor, and review history in a JSON database, scheduling reviews at optimal intervals based on the SM-2 algorithm and Cepeda et al. (2006) research. - Multi-Modal Vocabulary Introduction: Presents each new word with IPA pronunciation, mnemonics, collocations, word families, frequency rank, and contextual example sentences. - Exercise Generation: Creates seven exercise types including fill-in-the-blank, multiple choice, matching, sentence production, cloze tests, odd-word-out, and dictation. - Use Case: A Spanish learner asks to review vocabulary; the agent queries vocab.json for due words, presents randomized recall questions, updates SRS parameters based on self-assessed quality, and saves progress immediately. ## Quick Start Ask the agent to start a vocabulary review session for your target language, and it will pull all words due today from your personal vocabulary database.

Frequently Asked Questions about langtutor-vocab

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review vocabulary with spaced repetition?

Start a review session and the agent queries vocab.json for words where next_review is today or earlier. It presents randomized recall questions, asks you to rate recall quality (Easy, Good, Hard, Forgot), then updates intervals and ease factors using the SM-2 algorithm.

What is the SM-2 algorithm for flashcard scheduling?

SM-2 is a spaced repetition algorithm that adjusts review intervals based on recall quality. Correct answers multiply the interval by an ease factor, while forgotten words reset to a 1-day interval. This implementation adds modifications from the Cepeda et al. (2006) meta-analysis on optimal spacing.

How many new vocabulary words should I learn per session?

The agent limits new words to a maximum of 20 per session to respect cognitive load limits. It prioritizes high-frequency words first, since the top 1000 words in a language cover roughly 80% of everyday communication.

Where is my vocabulary progress data stored?

All vocabulary data is stored in a JSON file at Language-Tutor/learner-data/[language]/vocab.json. The agent writes updates immediately after each review and creates a timestamped backup before each write to prevent data loss.

What happens if my vocab.json file is corrupted or missing?

The agent first attempts recovery from the vocab.json.backup file. If recovery fails or no learner data exists, it directs you to the langtutor-assessor agent to complete an initial placement test and create a new learning plan.

Can I import my own vocabulary lists into the system?

Yes, the personal dictionary feature accepts CSV or JSON imports with word and translation columns, plus optional part of speech, examples, and frequency data. Imported words are filtered against your CEFR level and existing mastered vocabulary.