impasto-miniature-landscape
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of natural language instructions and metadata. No code execution, network requests, or sensitive file access patterns were found.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data in the form of user-provided photos (SKILL.md). It establishes boundary markers by instructing the agent to use the source photo only as a reference for subject, structure, pose, spatial relationships, and color, which mitigates the risk of the agent following instructions embedded within the images. The primary capabilities are image generation and editing (SKILL.md), and no explicit sanitization of visual data is performed.
Audit Metadata