asset-generation

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes instructions to execute npx @agent-native/core@latest for local customization and launching the application skill. This refers to the official toolchain associated with the service provider.
  • [PROMPT_INJECTION]: The skill architecture presents a surface for indirect prompt injection as it ingests user input and library metadata to drive generation tools.
  • Ingestion points: User prompts, composer mentions (@brand-kit, @preset), and collection style analysis in SKILL.md.
  • Boundary markers: Not explicitly defined in the workflow instructions.
  • Capability inventory: Includes sensitive generation actions like generate-image, generate-video, and export-asset across all sections of SKILL.md.
  • Sanitization: The instructions suggest conservative enhancement of prompts but do not specify formal input sanitization or validation protocols.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:51 AM
Security Audit — agent-trust-hub — asset-generation