The Agent Skills Directory

[SAFE]: No security issues detected. The skill follows standard practices for a content drafting tool.
[PROMPT_INJECTION]: The skill has a low-risk attack surface for indirect prompt injection. It ingests user data and reads local configuration files, but lacks the necessary capabilities (such as network access or file-writing) to enable data exfiltration or system damage.
Ingestion points: SKILL.md defines user inputs for topics and key messages, and references local settings files for brand voice.
Boundary markers: None explicitly defined.
Capability inventory: The skill is limited to Read, Glob, and Grep tools; it cannot perform network operations or write files.
Sanitization: Not explicitly defined.

draft-content