JSON is a text-based format for representing structured data with objects, arrays, values, and name-value pairs. In prompt-driven development, naming JSON explicitly helps an LLM return data in a predictable structure. section 2.4.3 section 6.1.1
Specify the expected structure before requesting generated data:
- Name JSON as the required output format. section 2.4.3
- Define fields, data types, allowed values, ranges, and nesting explicitly. section 6.1.1
- Request a fixed number of objects and check that the count is correct. section 6.1.1
Booking record
A JSON object can combine strings, numbers, booleans, and a nested booking_dates object containing checkin and checkout values. section 2.4.3
{
"firstname": "John",
"lastname": "Doe",
"totalprice": 150.50,
"deposit_paid": true,
"booking_dates": {
"checkin": "2023-08-01",
"checkout": "2023-08-10"
}
}
Inspect a compact JSON booking object with mixed value types and nested dates. section 2.4.3
