design-taste-frontend

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon untrusted data from user briefs and external URLs to influence its output.
  • Ingestion points: Section 0.A directs the agent to read signals including URLs and vibe words provided by the user to infer design directions.
  • Boundary markers: Absent. There are no instructions provided to ensure the agent ignores potential instructions embedded within the user-provided data or linked web content.
  • Capability inventory: The skill allows the agent to generate code, write files, and potentially execute package installation commands.
  • Sanitization: Absent. No sanitization or validation steps are included for processing external content.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of various third-party frontend libraries and design systems.
  • Evidence: Instructions include installing packages like Fluent UI, Material Web, Carbon, and Radix UI via standard package managers.
  • Note: These downloads target well-known organizations and official registries which are considered safe sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 01:02 AM
Security Audit — agent-trust-hub — design-taste-frontend