Generate JLPT N5 sentence building (starQuestion) questions in JSON format. Use this skill when the user asks to create, generate, or produce N5 sentence building questions, word-order exercises, or scrambled sentence questions. Output is a JSON object containing a "questions" array with starQuestion type items, each including sentencePrefix, sentenceSuffix, 4 scrambledWords, starPosition, and correctOrder. Trigger on requests like "N5 cümle birleştirme sorusu üret", "create N5 sentence building questions", "generate word order questions", "JLPT N5 mondai2 soru oluştur", or "scrambled sentence questions".
Generate N5-level sentence building (starQuestion) questions as JSON.
references/json-schema.md for field specifications.references/n5-grammar-points.md for word-order rules and scramble patterns.sentencePrefix + 4 words + sentenceSuffix.scrambledWords must be in a shuffled order (not the correct order).correctOrder as the array of indices mapping scrambled positions to the correct sequence.starPosition (0–3) indicating which slot in the correct order is the marked answer.id as n5_grammer_sentence_building_XXX where XXX is a zero-padded sequential number. Continue from the last existing ID in the file.backend/grammar/data/n5_sentence_building.json. Read the file first, then add new questions to the end of the questions array.jlpt-grammer-sentence-building-question-tester skill to validate the JSON file. If any questions fail, fix them and re-run the tester until all pass.python3 backend/grammar/scripts/upload_grammar_questions.py n5 --type sentence-buildingscrambledWords must NOT be in the correct order — always shuffle.correctOrder must be a valid permutation of [0, 1, 2, 3].