branded-graphics

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to fetch assets and copy from external brand websites, which introduces a potential surface for indirect prompt injection if those sources are compromised.
  • Ingestion points: Reads from brand's official sites for logos, palettes, and copy as defined in references/production-guide.md.
  • Boundary markers: None explicitly defined for external content, though the skill emphasizes strict adherence to supplied assets.
  • Capability inventory: The skill uses model calls (genmedia), media tools, and renderers (HTML/SVG) to create artifacts.
  • Sanitization: Relies on post-generation inspection by the agent rather than input sanitization or validation of the external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 07:17 AM