Security audit
Security checks for vulnerabilities and agentic risk
This is a local Chinese text-rewriting skill with no network, persistence, or privileged access, but it is explicitly aimed at removing AI-writing traces and should be used transparently.
Install only if you want a Chinese-focused rewriting helper. Use it for clarity, tone, and platform fit, and avoid using the AI-trace-removal features to misrepresent authorship, bypass disclosure rules, or violate academic, workplace, or platform policies.
The entire skill description, usage guidance, examples, and feature documentation are presented only in Chinese, with no indication that other languages are supported or that Chinese is a deliberate region-specific requirement. Under the policy, forcing a specific language without user opt-in is a natural-language policy concern.
The skill’s trigger phrases are broad enough to match routine editing requests like polishing, rewriting, or making text sound more natural. In an agentic system, this can cause over-selection of the skill and lead it to handle requests outside its intended niche, which increases the chance of unintended behavior such as unnecessary transformation of user content or evasion-oriented rewriting.
The usage conditions do not clearly separate 'humanization' from ordinary style editing, so the skill may be invoked for many common writing tasks. That ambiguity is risky because this skill is explicitly framed around removing 'AI traces,' which can be repurposed to obscure provenance or bypass AI-detection policies when used in broader contexts.
This markdown file presents all guidance in Chinese and does not indicate that users may select another language or that the skill is intentionally limited to a Chinese-language audience. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.
The script is explicitly designed to remove indicators of AI authorship and make output appear human-written, which can facilitate deception, policy evasion, academic dishonesty, spam, or undisclosed synthetic content. In this skill context, that behavior is the core advertised function, making the misuse risk more pronounced rather than incidental.
This Python skill is built entirely around Chinese replacements, Chinese filler phrases, and Chinese-only output embellishments such as prefixes and labels. Because the file contains no opt-in or language-selection mechanism, it effectively forces a specific language/locale behavior, which matches the natural-language policy violation criteria.
The scoring system and messages create an optimization loop that rewards reducing 'AI traces' and explicitly praises content as having removed AI indicators. That turns the tool from simple style editing into a concealment workflow, increasing its utility for deceptive rewriting and repeated refinement until authorship signals are minimized.
This markdown file contains user-facing instructional content exclusively in Chinese, and there is no indication that users can opt into another language or that the file is intentionally limited to a Chinese-speaking context. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern.
No suspicious patterns detected.
