Extracts fields from unstructured text according to a given JSON Schema, outputting only valid JSON and using null for missing fields.
Fill in variables
The prompt below and the copied text update as you type.
# Task
Extract information from the text below and output JSON that strictly conforms to the JSON Schema.
# JSON Schema
{{schema}}
# Text
"""
{{text}}
"""
# Rules
1. Output only JSON: no explanations, prefixes, suffixes or Markdown code fences.
2. Field names, types and enum values must match the schema exactly.
3. Use null for fields not found in the text; use [] for array fields not found.
4. Do not infer or fill in information that is not in the text.
5. Where the schema does not specify a format, write dates as YYYY-MM-DD and numbers without thousands separators.
6. When there are multiple entities, output them in the order they appear in the text.# 任务
从下方文本中抽取信息,输出严格符合 JSON Schema 的 JSON。
# JSON Schema
{{schema}}
# 文本
"""
{{text}}
"""
# 规则
1. 只输出 JSON,不要任何解释、前后缀或 Markdown 代码块标记。
2. 字段名、类型、枚举值必须与 Schema 完全一致。
3. 文本中找不到的字段填 null;找不到的数组字段填 []。
4. 不推测、不补全文本中没有的信息。
5. Schema 未规定格式时,日期写成 YYYY-MM-DD,数字去掉千分位分隔符。
6. 有多个实体时,按其在文本中出现的顺序输出。