auth-expert
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill uses 'grep' and 'find' to detect relevant libraries and files in the project directory. These are standard, read-only operations intended for environment detection and do not pose a security risk.\n- [CREDENTIALS_UNSAFE] (SAFE): No hardcoded secrets or API keys were found. The provided code templates correctly recommend using environment variables (e.g., 'process.env.ACCESS_TOKEN_SECRET') for sensitive configuration.\n- [DATA_EXFILTRATION] (SAFE): There are no network operations, 'curl'/'wget' commands, or external data transmission logic present in the skill.\n- [PROMPT_INJECTION] (SAFE): The skill contains no instructions aimed at bypassing AI safety filters, overriding system prompts, or extracting internal instructions.\n- [REMOTE_CODE_EXECUTION] (SAFE): No remote script downloads or dynamic execution patterns were identified.
Audit Metadata