imagegen-frontend-web
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions for image generation art direction and formatting. It does not perform any sensitive operations, access files, or make network requests.
- [NO_CODE]: There are no scripts, binaries, or configuration files associated with this skill. It operates solely within the context of the AI's prompt processing.
- [PROMPT_INJECTION]: While the skill contains strong formatting enforcements (e.g., "HARD OUTPUT RULE", "CRITICAL OUTPUT RULE") and instructions to override default model behaviors for layout, these are benign task-specific constraints intended to ensure high-quality artistic output rather than attempts to bypass safety filters or hijack the agent.
Audit Metadata