test_data_prompts.txt

1. Генерация синтетических валидных данных:
You are a test data generator for QA automation.

Generate a synthetic dataset of 20 users.

Rules:
- All data must be fictional.
- Do NOT use any real people or real companies.
- Emails must use example domains (example.com, test.com).
- Phone numbers must be format-valid but not real.
- Names must be fictional.
- Ages between 18 and 65.

Return result as a JSON array.

Each user must contain:
- id (uuid)
- firstName
- lastName
- email
- phone
- age
- country
- role (USER, ADMIN, MANAGER)

2. Генерация бизнес-реалистичных профилей:
You are generating business-realistic synthetic users for an online banking system.

Constraints:
- All data must be synthetic.
- No real PII.
- Each user must have:
  - profile
  - bankAccount
  - balance
  - status (ACTIVE, BLOCKED, PENDING)

Rules:
- BLOCKED users must have balance = 0
- PENDING users must not have a bank account number
- ACTIVE users must have realistic balance (100–50,000)

Generate 15 users.

Return structured JSON.

3. Генерация пограничных и невалидных данных:
You are a QA engineer generating negative and edge-case test data.

Generate a JSON dataset for user registration.

Include:
- 5 valid users
- 5 invalid users
- 5 edge cases

Invalid examples must include:
- wrong email formats
- too short and too long names
- invalid phone numbers
- missing required fields

Edge cases must include:
- maximum allowed field lengths
- unicode characters
- boundary ages (18, 65)

All data must be fictional.
Return JSON grouped by type: valid, invalid, edge.

4. Генерация данных строго по контракту (schema-driven)
You are a synthetic test data generator.

Generate 30 records that strictly follow this JSON Schema:

{
  "id": "uuid",
  "email": "string, email format, example domains only",
  "password": "8-20 chars, letters and digits",
  "age": "integer, 18-65",
  "marketingConsent": "boolean",
  "createdAt": "ISO 8601 datetime"
}

Rules:
- No real personal data.
- Emails must use example.com or test.com only.
- createdAt must be within last 90 days.

Return ONLY valid JSON.

5. Массовая генерация под performance
You are generating large-scale synthetic data for performance testing.

Generate a dataset of 1,000 fictional users.

Each user must have:
- userId
- email
- password
- accountStatus
- region
- registrationDate

Rules:
- No real data.
- Use only synthetic names and emails.
- Provide realistic distribution:
  - 70% ACTIVE
  - 20% PENDING
  - 10% BLOCKED

Output as downloadable-ready JSON structure.

6. Маскирование существующих данных (без раскрытия PII)
You are a data-masking engine.

I will provide a dataset.

Your task:
- Replace all personal data with synthetic equivalents.
- Keep formats and uniqueness.
- Preserve relationships between records.
- Do NOT output original values.

Mask:
- names
- emails
- phone numbers
- document numbers

Return masked JSON only.

7. Генерация data-factory под автотесты:
You are a Senior QA Automation Engineer.

Based on this user JSON structure, generate a Java test data factory.

Factory must:
- Generate valid users
- Generate invalid users
- Allow overriding any field
- Support random and fixed seed generation

Return:
- Java class
- Example usage in test

8. Контрольный промт (универсальный шаблон):
You are a professional QA test data generator.

Domain: {{SYSTEM_NAME}}

Data type: {{ENTITY}}

Constraints:
- No real PII.
- Must follow this schema: {{SCHEMA}}
- Must include: valid, invalid, and edge cases.
- Dataset size: {{N}} records.

Additional business rules:
{{RULES}}

Return JSON only.