The Agent Skills Directory

[SAFE]: The skill consists of markdown guidelines and evaluation data. No executable code, automation scripts, or binary files were found within the skill package. The instructions to read local context files like .agents/product-marketing.md are standard for agent personalization and do not involve access to sensitive credentials.
[PROMPT_INJECTION]: The skill is designed to ingest and analyze user-supplied data regarding business performance. While this represents an indirect injection surface, the skill lacks high-risk capabilities to exploit such an injection. * Ingestion points: User conversation and local context files (SKILL.md). * Boundary markers: None present. * Capability inventory: No internal command execution or network capabilities; relies on external skills for specific actions. * Sanitization: None present.

churn-prevention