Miami
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is comprised entirely of static Markdown files containing information about Miami's neighborhoods, food, transport, and safety. There are no executable scripts (.py, .js, .sh) or binary files included.
- [PROMPT_INJECTION]: No prompt injection patterns, role-play instructions, or attempts to override system constraints were found in the instructional content.
- [DATA_EXFILTRATION]: The skill does not request any network-enabled tools or include commands that could exfiltrate sensitive data. All information provided is static and intended for user guidance.
- [COMMAND_EXECUTION]: There is no usage of the dynamic context injection syntax (
!command) or any other method for executing shell commands. - [NO_CODE]: As a data-only skill, it does not possess the capability to perform dynamic execution or establish persistence mechanisms.
Audit Metadata