Security audit
Security checks for vulnerabilities and agentic risk
This is a content-writing skill with no executable code, but users should be aware it may analyze content samples or performance information they provide.
Before installing, understand that this skill is intended for drafting, editing, repurposing, and planning content. Share past posts, drafts, analytics, or audience data only when you want the agent to use them, and avoid providing sensitive or private material unless it is necessary for the task.
The skill description is broad enough to plausibly activate in many content-related contexts without clear boundaries on when it should or should not be used. Overly generic activation language can cause unintended invocation, leading the agent to apply the skill to sensitive inputs or tasks the user did not explicitly intend, which increases the chance of privacy, scope, and consent issues.
The skill states that it learns from previously published content and tracks what performs, but the description does not clearly warn users that their past content and performance data may be analyzed. Without explicit disclosure and consent language, users may unknowingly provide sensitive writing samples, audience data, or account metrics, creating privacy and expectation-of-use risks.
No suspicious patterns detected.
