ai-ai-testing
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill contains phrases such as 'Ignore previous instructions' and 'DAN' in
SKILL.mdandreferences/ai-testing-advanced.md. These are part of an attack taxonomy and safety testing guide intended to teach users how to evaluate model robustness against such attacks. They do not constitute an attempt to override the agent's behavior. - [EXTERNAL_DOWNLOADS]: The skill references various standard AI testing and evaluation libraries (e.g.,
deepeval,promptfoo,ragas,giskard,sentence-transformers). These are well-known tools in the AI development ecosystem. - [DATA_EXFILTRATION]: There is no evidence of unauthorized data access or exfiltration. The YAML configuration examples for CI/CD pipelines correctly demonstrate using secret managers (e.g.,
${{ secrets.OPENAI_API_KEY }}) rather than hardcoding credentials. - [COMMAND_EXECUTION]: The skill includes Python code snippets for building testing runners. These scripts perform standard data processing, statistical analysis, and API calls to LLM providers for evaluation purposes.
Audit Metadata