shaman-einstein
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill uses high-intensity roleplay instructions to establish a specific 'Shaman Einstein' persona. It includes directives for the agent to adopt a hostile or 'scolding' tone (referred to as a 'roaring system' and 'intellectual humiliation') when it encounters what it deems 'unimaginative' input. However, these are character traits for roleplay and do not attempt to bypass safety guidelines, escape the sandbox, or override the agent's core instructions.
- [DATA_EXFILTRATION]: No network requests, API calls, or attempts to access sensitive files (such as environment variables or credentials) were found.
- [COMMAND_EXECUTION]: The files do not contain any shell commands, script execution, or system-level operations.
- [EXTERNAL_DOWNLOADS]: No external resources, remote scripts, or third-party packages are referenced or downloaded.
Audit Metadata