sd-xuanhuan-engine
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists exclusively of markdown documentation and reference libraries. It does not include any executable scripts, binaries, or configuration files that could trigger code execution.
- [PROMPT_INJECTION]: The instructions contain 'Yielding Clauses' (让行条款) which instruct the AI to prioritize the skill's specific Xuanhuan rules over generic combat guidelines. These are architectural instructions designed to ensure the AI adheres to a specific artistic style and do not represent a bypass of safety filters or ethical guardrails.
- [SAFE]: The skill does not attempt to access sensitive files, perform network operations, or utilize obfuscation techniques. All references to external platforms (Seedance, H3) and internal tools (validate_prompt.py) are consistent with the stated purpose of professional video prompt engineering.
Audit Metadata