ab-test-setup
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of markdown-based guidelines for statistical experimentation and hypothesis testing. There are no malicious patterns detected.
- [NO_CODE]: No executable scripts, shell commands, or automation logic are present in the skill files. It functions purely as a prompt-based framework for human-AI interaction.
- [DATA_EXFILTRATION]: No network access or data transmission capabilities are defined. The skill does not interact with external APIs or remote servers.
- [PROMPT_INJECTION]: The instructions do not contain markers for overriding system prompts, bypassing safety filters, or extracting sensitive agent configurations.
Audit Metadata