Security audit
Security checks for vulnerabilities and agentic risk
This is a writing-only skill for turning long-form content into social posts, with no executable code, persistence, credential access, or hidden data handling.
This skill appears safe to install for content marketing workflows. Users should still review generated posts before publishing, especially because the skill may activate for broad social-media or thread-writing requests.
The trigger phrases are broad and overlap with common user requests, which can cause this skill to activate in situations where the user did not explicitly intend to invoke it. That increases the chance of prompt/skill hijacking, incorrect tool routing, or unintended transformation of user content, especially in systems where multiple skills compete for activation.
The activation conditions describe when the skill should activate but do not define exclusion boundaries, causing ambiguous matching on generic content-rewriting requests. In an agent environment, this can lead to unintended execution, override of more appropriate skills, or processing of sensitive content under the wrong persona or workflow.
No suspicious patterns detected.
