mobile-architecture

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely composed of markdown instructions and templates, with no executable code or scripts included.
  • [PROMPT_INJECTION]: Analysis of the instructions and playbooks revealed no attempts to bypass safety filters or override the agent's core instructions.
  • [DATA_EXFILTRATION]: The skill does not request access to sensitive credentials or system files, nor does it perform any network requests.
  • [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or dynamic execution mechanisms present in the skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 08:35 AM
Security Audit — agent-trust-hub — mobile-architecture