The Agent Skills Directory

[SAFE]: The skill provides structured guidelines for generating marketing content. A thorough review of the instructions and reference templates reveals no malicious patterns, prompt injection attempts, or unauthorized system operations.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill instructions include a check for local product marketing context files (e.g., .agents/product-marketing.md). This is a standard practice for context-aware agents to personalize output and does not involve accessing sensitive system credentials or private user data.
[INDIRECT_PROMPT_INJECTION]: The skill defines a clear ingestion surface by reading local marketing documentation. As these files are intended to be part of the agent's internal knowledge base and the skill lacks dangerous capabilities like network exfiltration or command execution, the risk of indirect injection is negligible.

competitors