Security audit
Security checks for vulnerabilities and agentic risk
This skill is meant to post to X, but its trigger and credential/posting flow are too loosely scoped for a public posting action.
Review this carefully before installing. Only use it with an explicit posting request, do not provide unnecessary secrets, and require the agent to show the exact post content and receive confirmation before publishing to X.
The trigger description uses broad phrasing like '或类似', which can cause the assistant to invoke the posting skill in ambiguous situations. In a skill that can publish content to an external social platform, vague triggering increases the risk of unintended posting or accidental credential-handling flows.
This skill handles outbound transmission of user-provided content and API credentials to X-related tooling, but the description does not warn users that their text, attachments, and authentication material may be sent to external APIs. Without a clear disclosure, users may unknowingly expose sensitive content or secrets, especially because social posting is irreversible once published.
The skill documentation is entirely in Chinese and does not indicate that users may choose another language or locale. If this skill is intended for broader use, that can constitute a language-policy issue because the locale is effectively fixed without an explicit opt-in or justification.
No suspicious patterns detected.
