The Agent Skills Directory

[SAFE]: No security issues detected. The skill purely provides instructional content and design specifications.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential injection surface by instructing the agent to read .claude/product-marketing-context.md in SKILL.md (Ingestion point). The instructions do not specify boundary markers or delimiters for this input (Boundary markers). However, the skill contains no dangerous capabilities such as file writing, command execution, or network access across SKILL.md (Capability inventory), and no specific sanitization steps are defined (Sanitization).

brand-guidelines