experimentation-and-ab-testing

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [PROMPT_INJECTION]: No attempts to override agent behavior or bypass safety guidelines were found. The instructions emphasize scientific rigor and integrity.
  • [DATA_EXFILTRATION]: No network access, credential harvesting, or sensitive file operations are performed or requested.
  • [REMOTE_CODE_EXECUTION]: The skill consists entirely of markdown documentation and JSON test cases with no executable code or remote dependency downloads.
  • [OBFUSCATION]: No hidden characters, encoded strings, or homoglyphs were identified in any of the analyzed files.
  • [INDIRECT_PROMPT_INJECTION]: While the skill ingests brand profiles and goals, it does not interpolate this data into executable contexts or demonstrate specific injection vulnerabilities.
  • [SAFE]: All instructions and examples align with the stated purpose of A/B testing design, and the scope correctly identifies platform limitations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 02:27 PM
Security Audit — agent-trust-hub — experimentation-and-ab-testing