Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent recipe-generation skill with disclosed paid access, though users should verify the external payment helper before running it.
Before installing, confirm you trust the payment helper package and prefer a pinned version or official provenance if available. Expect the skill to send recipe input to the listed paid endpoint when used.
The skill display name and the entire instruction content are presented only in Chinese, with no indication that users may choose another language or locale. Under the policy, language constraints should be optional or clearly justified rather than implicitly enforced.
The description lists activation phrases such as "菜谱", "怎么做", and "今天做什么菜", which are common conversational expressions and not narrowly scoped to explicit skill invocation. The file also does not provide exclusion conditions or negative examples to clarify when these phrases should not trigger the skill.
The skill instructs users to run an unpinned package via npx ...@latest, which introduces a supply-chain risk: future package updates or a compromised publisher account could cause unexpected code to execute on the buyer side. This is more concerning because the command installs and runs code locally, outside the skill sandbox, even though the skill claims the script only orchestrates payment.
No suspicious patterns detected.
