art-3d
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate instructions for generating image prompts and contains no attempts to bypass safety filters or override agent behavior.\n- [SAFE]: No network activity, sensitive file access, or hardcoded credentials were detected in the skill or its references.\n- [SAFE]: The agent's capabilities are restricted to the Read tool for local reference files, with no evidence of remote code execution or persistence mechanisms.\n- [PROMPT_INJECTION]: Evaluation of the indirect prompt injection surface:\n
- Ingestion points: The skill processes a 'character or subject' argument in SKILL.md.\n
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present for the user-supplied argument.\n
- Capability inventory: The agent is limited to the Read tool and text prompt generation.\n
- Sanitization: No input validation or filtering is defined.\n
- Conclusion: The surface is safe as there are no exploitable high-privilege capabilities such as network access or file system writes.
Audit Metadata