The Agent Skills Directory

[DATA_EXPOSURE]: The skill is designed to read local project documentation, such as files in .agents/SYSTEM/ARCHITECTURE.md and .agents/SOP/, to identify brand voice and context. This data is used within the agent's context to ensure content consistency and is not exfiltrated to any external servers.
[EXTERNAL_DOWNLOADS]: The skill suggests the use of an external plugin from a third-party GitHub repository (coreyhaines31/marketingskills) for additional marketing tactics. This is a recommendation for the user and does not involve the automatic execution of remote code by the skill itself.
[PROMPT_INJECTION]: The skill processes project-specific data like README files and previous social media posts which could potentially contain indirect instructions. However, because the skill does not possess dangerous capabilities such as arbitrary command execution or network access, the risk associated with this ingestion surface is negligible.

content-creator