Security audit
Security checks for vulnerabilities and agentic risk
The skill is mostly coherent, but it can automatically publish WeChat public-account articles after an IP check without a clear confirmation step or narrow activation scope.
Review this before installing if you use WeChat publishing in a real account. The IP check itself is straightforward, but publication can happen automatically when the stored IP matches, so use it only if you are comfortable with that workflow and with contacting ip38.com to reveal your public IP lookup metadata. Narrowing triggers and requiring explicit confirmation before every publish would reduce the risk.
The entire skill description, examples, and interaction instructions are presented only in Chinese, with no indication that users may choose another language or that the skill is intentionally restricted to Chinese-language use. This can violate language/locale policy when no opt-in or documented justification is provided.
The skill explicitly instructs use of a third-party website to discover the user's public IP but does not warn that this sends network metadata to an external service. That omission can expose the user's IP address, access timing, and possibly browser fingerprinting data without informed consent, which is a real privacy risk even if the site is legitimate.
The trigger list includes very broad phrases such as '检查 IP', 'wechat ip', and '发布公众号' that can match ordinary user requests not clearly limited to this specific skill. Overbroad activation can cause unintended invocation, letting the skill intercept unrelated prompts and potentially perform actions or expose capabilities outside the user's intent.
The natural-language fields force a Chinese-language interaction model through the description and activation phrases, but there is no indication that this is optional or limited to a Chinese-speaking context. That can violate language-choice policy when a skill does not document user opt-in or locale restriction.
No suspicious patterns detected.
