jlpt-n5-grammer-text-integration-question-tester - SKILL.md Agent Skill

name: jlpt-n5-grammer-text-integration-question-tester description: Validate JLPT text-level grammar (Mondai 3) JSON files produced by jlpt-n5-grammer-text-integration-question-creator. Use this skill when the user asks to test, validate, check, or verify Mondai 3 (text integration) questions in JSON format. Checks (1) schema correctness for "textFlow" type, (2) blankNumber/position sequencing, (3) correctAnswer is valid, (4) level is n5. Trigger on requests like "Mondai 3 soruları test et", "validate text integration questions", "textFlow JSON doğrula", or after generating questions with the text integration creator skill.

JLPT Text Integration Question Tester

Validate JLPT N5 text-level grammar (Mondai 3) JSON using scripts/validate_text_integration.py.

Workflow

Identify the JSON to validate — either a file path provided by the user, or JSON content in the conversation.
If JSON is in context (not a file), write it to a temporary file first, then run the script.
Run the validator:

# From a file
python3 <skill_dir>/scripts/validate_text_integration.py <path/to/questions.json>

# From stdin (JSON in context)
python3 <skill_dir>/scripts/validate_text_integration.py --stdin <<'EOF'
{ ...json... }
EOF

If the script reports failures, summarize them and stop here.
Linguistic review: If all questions pass the script, read the JSON file and verify each question linguistically:
- Reconstruct the full text by joining textSegments with the correct options[correctAnswer] in between.
- Verify the reconstructed text is a coherent, grammatically correct N5-level narrative.
- Verify each correct answer is the ONLY correct option among the 4 choices for that blank.
- Verify all distractors are plausible yet incorrect in the context of the narrative.
- Verify all vocabulary and kanji are within N5 scope.
- Report any linguistically incorrect questions with an explanation.
Report final results — script validation + linguistic review summary.

What the script checks per question (mechanical)

Check	Rule
`id` format	Must match `n5_grammer_text_integration_XXX`
`level`	Must be `"n5"`
`type`	Must be `"textFlow"`
`title`	Must be a non-empty string
`textSegments`	Array of strings (at least 2)
`blanks`	Array of objects; length must be `len(textSegments) - 1`
`blankNumber`	Positive integer
`position`	Must match the index in the `blanks` array
`options`	Must be an array of exactly 4 non-empty strings
`correctAnswer`	Must be an integer 0–3

Output

The script prints a report with ✅ PASS or ❌ FAIL per question and a summary line. Exit code 0 = all passed, 1 = one or more failed.