carplay

Fail

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: CRITICALNO_CODE
Full Analysis
  • SAFE (SAFE): Detailed review of the 7 Markdown files confirms they contain only technical documentation and API references for the CarPlay framework.
  • NO_CODE (SAFE): There are no scripts, binaries, or configuration files included in the skill that could be executed by an agent.
  • DATA_EXFILTRATION (SAFE): No functions for reading sensitive local files or transmitting information to external servers were identified.
  • EXTERNAL_DOWNLOADS (SAFE): The files do not contain any links or commands (e.g., curl, wget) to download or execute external resources.
  • PROMPT_INJECTION (SAFE): The content is descriptive and does not contain instructions aimed at overriding agent behavior or bypassing safety filters.
Recommendations
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 19, 2026, 05:12 PM