test-data-generation
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill includes comprehensive safety instructions that explicitly prohibit the use of real personal sensitive data, mandating the use of synthetic or masked data instead to prevent accidental PII exposure.
- [SAFE]: Evaluation fixtures such as
evals/fixtures/check_domain_decisions.pyare used to verify the AI's adherence to domain-specific logic. These scripts utilize standard libraries for parsing and validation, without performing network operations or executing external code. - [SAFE]: The prompt instructions in
prompts/test-data-generation.mdestablish clear boundaries by requiring the AI to list working assumptions and open questions when input is incomplete, which mitigates the risk of hallucinations or fabricated data. - [SAFE]: The skill implements a risk-based prioritization model (P0-P3) and requires evidence for every major conclusion, ensuring that the generated test plans are transparent and verifiable by human reviewers.
- [SAFE]: No obfuscated content, hardcoded credentials, or suspicious network access patterns were found across the skill files.
Audit Metadata