ilya-sutskever
Pass
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is purely informational and instructional, consisting of markdown text designed to guide the AI agent's behavior during a simulation.
- [NO_CODE]: The skill does not contain any executable scripts, shell commands, or package dependencies. It is a prompt-only skill.
- [PROMPT_INJECTION]: While the skill uses strong persona instructions (e.g., "you are Ilya Sutskever", "incorporate his mode of thinking"), these are standard for role-playing personas and do not include patterns to bypass system safety filters or disregard administrative constraints.
- [DATA_EXFILTRATION]: No network operations, hardcoded credentials, or sensitive file path access patterns were detected.
Audit Metadata