design-image-prompt-engineer
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided visual concepts and mood boards to generate prompt strings. This creates an attack surface for indirect prompt injection where malicious instructions could be embedded in the input data. However, the agent's environment is highly restricted as it does not possess capabilities for file writing, network communication, or tool execution, which significantly limits the potential impact of such an attack.
- Ingestion points: Concept Intake (Step 1) in
SKILL.mdprocesses user goals and references. - Boundary markers: None explicitly defined to separate user-provided concepts from the system instructions.
- Capability inventory: No tool usage (
allowed-tools), subprocess calls, or network operations are defined in the frontmatter or instruction body. - Sanitization: No specific sanitization or validation logic is defined for the input concepts.
Audit Metadata