Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only writing-style skill with broad and sometimes rigid editing guidance, but no code execution, persistence, credential access, or hidden data flow.
Install this only if you want an assertive anti-AI-style editor. It may over-normalize prose or conflict with creative voice, formal tone, accessibility needs, or user-provided style constraints, so invoke it deliberately for anti-slop cleanup rather than all writing tasks.
The skill description is broad enough to trigger on generic drafting, editing, reviewing, or scoring tasks, which can cause the agent to apply this skill outside the user's intended style constraints. That creates a prompt-scope risk: ordinary writing requests may be silently rewritten according to this skill's rigid anti-style rules, degrading fidelity to user intent across many contexts.
The skill mandates absolute stylistic rules such as removing all adverbs, banning passive voice, and forbidding certain structures without requiring user opt-in or contextual exceptions. In an agent setting, this can override the requested tone, genre, or linguistic needs, causing unwanted transformation of user content and making outputs less accurate, less inclusive, or less suitable for the task.
The title '中文AI套话与填充词黑名单' establishes a language-specific rule set for Chinese output, but the file does not indicate that this is optional, user-selected, or limited to a justified regional context. Under the policy, forcing a specific language or locale without opt-in is a natural-language policy violation.
This markdown file presents all instructions and examples exclusively in Chinese, and there is no indication that the user can choose another language or that the Chinese-only scope is required. Under the policy rule for natural-language constraints, forcing a specific language without opt-in is a locale/language policy concern.
The title '中文AI写作结构模式' presents the skill as specifically Chinese-language guidance, and the document contains no indication that users may choose another language or locale. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern.
No suspicious patterns detected.
