Security audit
Security checks for vulnerabilities and agentic risk
This is a text-only LOFTER fanfiction writing skill with no code execution, network access, persistence, credential use, or destructive behavior.
Install this if you want a Chinese LOFTER-style fandom writing workflow and are comfortable with bundled fanfiction samples influencing tone and structure. Review generated content for fandom appropriateness, rating boundaries, and sensitive tropes before posting.
Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
# Title Patterns
Use these title families when stronger click-through is needed.
Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
# Title Patterns
Use these title families when stronger click-through is needed.
Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
# Title Patterns
Use these title families when stronger click-through is needed.
Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
# Title Patterns
Use these title families when stronger click-through is needed.
The skill explicitly mandates Chinese prose output ('Write in readable, emotional, screenshot-friendly Chinese') without indicating this is conditional on the user's language or a hard platform requirement. This can override user preferences, reduce usability, and cause unintended disclosure or mishandling when downstream systems assume language neutrality.
The skill is described as a broad end-to-end writing workflow with no clear activation boundaries, narrow domain constraints, or explicit limits on when it should engage. Overly generic triggers increase the chance the agent will invoke this skill for unrelated writing requests, causing prompt-scope overreach, unintended instruction capture, or misuse in contexts the skill was not designed to handle.
This plain-text skill file presents all natural-language content in a single language and does not indicate that Chinese is an intentional, region-specific requirement. Under the policy rule, forcing a specific language without user opt-in is a natural-language policy violation.
The line includes the statement “男人和男人也能接吻吗?” framed as shocked objection, which reads as discriminatory natural-language content about same-sex affection rather than a neutral description. SQP-3 applies to all file types and covers policy-violating language in natural-language text.
The file is entirely written as a Chinese-roleplay instruction set and immediately imposes a Chinese persona and cultural framing on the assistant without any opt-in or alternative language/locale path. Under the stated policy, language or locale constraints should not be forced unless the skill clearly offers user choice or documents a justified region-specific scope.
Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
遇事不决，理性至上。
No suspicious patterns detected.