creative-eye
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is comprised entirely of instructional content, workflow frameworks, and evaluation prompts. No executable code (.py, .js, .sh), binary files, or network-bound operations were detected in the skill files.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to have an AI agent ingest and analyze external visual content such as reference images and brand materials. While this establishes an ingestion point for untrusted data, the skill possesses no dangerous capabilities that could be exploited through this surface.
- Ingestion points: External reference images and brand study materials as described in SKILL.md.
- Boundary markers: No explicit boundary markers or 'ignore' instructions are provided for the vision model prompts.
- Capability inventory: No capabilities for subprocess calls, code execution (eval/exec), network operations, or file system modifications were found.
- Sanitization: No automated sanitization is present as the skill operates via natural language prompts without technical data processing.
Audit Metadata