anime-to-life
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [PROMPT_INJECTION] (SAFE): The instructions are descriptive and do not attempt to bypass AI safety filters, ignore previous instructions, or extract system prompts.
- [DATA_EXFILTRATION] (SAFE): There are no commands or scripts that access local files or transmit data to external servers.
- [REMOTE_CODE_EXECUTION] (SAFE): The skill does not include any scripts, package dependencies, or remote download commands.
- [INDIRECT_PROMPT_INJECTION] (LOW):
- Ingestion points: The skill processes user-uploaded images and suggests using search results to identify characters.
- Boundary markers: None present.
- Capability inventory: No code execution or file-system capabilities.
- Sanitization: Not applicable as there is no code. The risk is limited to the AI's interpretation of external data for image generation parameters.
Audit Metadata