send-experiment-designer
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a local Python script (
experiment.py) to perform statistical calculations. This is a functional requirement to process experimental data and is executed within the plugin's local environment using standard CLI arguments.- [INDIRECT_PROMPT_INJECTION]: The skill processes external data exported from marketing platforms. It mitigates potential injection risks by explicitly instructing the agent to interpret this data strictly as values and not as instructions, adhering to defined security practices for handling external content.
Audit Metadata