key-visual-design

Pass

Audited by Gen Agent Trust Hub on Sep 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or suspicious code behaviors were identified within the skill instructions or reference materials.
  • [EXTERNAL_DOWNLOADS]: The skill references portfolio sites for well-known design firms, including Pentagram, Studio Dumbar, and Studio Moross, to provide real-world examples of campaign design principles.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest project-specific information such as design briefs, creating a potential surface for indirect injection.
  • Ingestion points: The workflow in SKILL.md utilizes external data from "current brief" and "campaign truth" inputs provided by the user.
  • Boundary markers: No explicit delimiters or instructions are used to separate untrusted brief content from agent instructions.
  • Capability inventory: None. The skill does not invoke any system commands, write to the filesystem, or perform network operations beyond reading the provided reference links.
  • Sanitization: Not present, as the skill lacks any executable capabilities that would require input sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 27, 2026, 12:49 PM