mobile-app

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses local HTML and CSS templates (assets/template.html, references/layouts.md) to generate UI mockups. It does not execute external scripts, install packages, or perform network requests.
  • [SAFE]: Instructions in references/checklist.md explicitly prohibit the use of external image URLs, requiring the use of local CSS placeholders instead. This is a security-positive rule that prevents potential data exfiltration or tracking via remote assets.
  • [PROMPT_INJECTION]: While the skill ingests untrusted data from a user's 'brief' to populate the UI templates (Archetypes A-F), the risk is minimal as the output is restricted to an HTML/CSS artifact for design previews.
  • Ingestion points: User-provided brief content is interpolated into bracketed text within layouts.md.
  • Boundary markers: The skill relies on structured templates but does not include specific 'ignore embedded instructions' warnings for the brief content.
  • Capability inventory: The skill is limited to generating text/html artifacts; it has no access to file writing, subprocesses, or environment variables.
  • Sanitization: No explicit sanitization of the brief copy is performed, but the restricted output format mitigates most indirect injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 10:27 AM
Security Audit — agent-trust-hub — mobile-app