access-control-roles
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [PROMPT_INJECTION]: No attempts to override system instructions or bypass safety filters were detected.\n- [DATA_EXFILTRATION]: No credential leaks or unauthorized data transmission patterns were found. The skill explicitly warns the user against putting sensitive mandate secrets into unchecked systems.\n- [EXTERNAL_DOWNLOADS]: The skill references trusted official legal portals (gesetze-im-internet.de, dejure.org, eur-lex.europa.eu) for information verification, which is appropriate for its stated purpose.\n- [REMOTE_CODE_EXECUTION]: No remote code execution patterns or package installations were found.\n- [OBFUSCATION]: The content is clear markdown without any hidden characters or encoded strings.\n- [NO_CODE]: This skill consists entirely of natural language instructions and does not include any scripts or executable code.
Audit Metadata