nano-banana
Warn
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: Instructions in
SKILL.mdprovide shell command templates that incorporate user-provided prompts directly into a command line (e.g.,python3 gen.py "prompt"). This creates a vulnerability to command injection where an attacker could use shell metacharacters like backticks or semicolons to execute arbitrary code on the host system. - [COMMAND_EXECUTION]: The skill instructs the agent to append environment variables to
/etc/profile, a system-wide configuration file. Modifying system profiles is a sensitive operation that can persist settings across all users and potentially lead to environment manipulation.
Audit Metadata