Security audit
Security checks for vulnerabilities and agentic risk
This is a text-editing skill that rewrites user-facing copy to sound less AI-generated, with no code execution, persistence, credential access, or hidden data flow.
Install this if you want an aggressive copy-editing assistant for removing AI-like tone. Be aware that its triggers are broad, so it may rewrite normal marketing, email, or public-facing copy whenever the agent determines that copy will be read by real users.
Several trigger phrases like 'humanize this,' 'make it sound natural,' and similar wording are common editing requests that may appear in ordinary workflows unrelated to this specific skill. Because they are generic, they can spuriously activate the skill and cause unrequested rewriting behavior, reducing predictability and potentially overriding user intent or other specialized instructions.
Several trigger phrases like 'humanize this,' 'make it sound natural,' and similar wording are common editing requests that may appear in ordinary workflows unrelated to this specific skill. Because they are generic, they can spuriously activate the skill and cause unrequested rewriting behavior, reducing predictability and potentially overriding user intent or other specialized instructions.
This file is a natural-language instruction guide explicitly scoped as '日本語' and prescribes Japanese writing style rules without offering any language or locale choice. Under the policy rule, forcing a specific language is a violation unless the locale constraint is clearly documented and justified as region-specific.
This markdown file includes natural-language instructions that specifically direct behavior for '收到中文文案修改任务时', effectively constraining operation to Chinese without any user opt-in or explanation of why a language restriction is required. Under the policy rule, forcing a specific language can be a locale-policy violation unless the constraint is explicitly justified or optional.
No suspicious patterns detected.
