Security audit
Security checks for vulnerabilities and agentic risk
This is a static Seedance prompt-writing skill, but it includes guidance for disguising restricted IP and moderation-sensitive terms, so it should be reviewed before installation.
Review before installing. The skill is not trying to access your computer or credentials, but it may steer broad video-prompt requests into Seedance-specific behavior and includes instructions that could be used to bypass platform moderation or IP restrictions. Install only if you are comfortable editing or removing those sections and keeping use within the target platform's rules.
The platform access section hardcodes specific Chinese-language services and URLs without any user opt-in or locale check, which can force a regional workflow the user did not request. This is risky because it may direct users to third-party platforms inappropriate for their jurisdiction, language, or trust expectations, and can override user autonomy in tool selection.
The trigger list is overly broad for a generic content-generation skill, including common phrases like 'video prompt' and 'AI video' that could cause the skill to activate for many unrelated requests. This creates prompt-routing risk: users may be steered into this skill unnecessarily, increasing exposure to its policy-evasion guidance and reducing user control over which behavior is invoked.
This section explicitly teaches users to replace banned proprietary and moderation-sensitive terms with descriptive paraphrases in order to avoid platform rejection. That is a classic moderation-evasion pattern: it preserves the disallowed semantic content while disguising the keywords, enabling users to generate content tied to protected IP or restricted topics despite stated platform rules.
No suspicious patterns detected.
