viral-generator-builder

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): The content consists of architectural patterns and role-play instructions for a 'Viral Generator Architect'. There are no instructions to ignore system rules or bypass safety protocols.
  • [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths, or network operations to external domains were found.
  • [Indirect Prompt Injection] (LOW): The skill describes tools that ingest untrusted user data (names, quiz answers, photos).
  • Ingestion points: Input Design section in SKILL.md (Name, Birthday, Quiz answers, Photo upload).
  • Boundary markers: Absent in the provided code templates.
  • Capability inventory: No system-level capabilities or network operations are included in the skill scripts.
  • Sanitization: Not mentioned in the patterns, which is a common omission in design templates but presents no direct risk to the agent executing the skill.
  • [Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill does not reference or download any external packages or remote scripts.
  • [Dynamic Execution] (SAFE): Contains static JavaScript snippets for scoring and hashing logic. No use of eval(), exec(), or runtime compilation was detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:41 PM