evaluating-new-technology
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- SAFE (SAFE): A comprehensive scan of all files confirms the absence of any malicious patterns or security risks across the 10 monitored threat categories.
- NO_CODE (SAFE): The skill is purely instructional and contains no Python, JavaScript, or shell scripts, nor does it include package management files.
- DATA_EXPOSURE (SAFE): The instructions explicitly state 'Do not request secrets' and emphasize evaluating technology within the context of security, privacy, and compliance constraints.
- PROMPT_INJECTION (SAFE): No instructions were found that attempt to bypass safety filters or override the agent's core operational logic.
Audit Metadata