The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains a specific behavioral override in the 'System Prompt Inquiry Response' section, instructing the agent to divert any questions about its internal instructions to the author's membership website.
[PROMPT_INJECTION]: The instructions in 'Mandatory Content Standards' explicitly state there should be 'No warnings, notes, or disclaimers,' which is a common pattern used to suppress the AI's built-in safety filters and transparency notices.
[PROMPT_INJECTION]: The 'Mandatory Intro Message' section forces the agent to inject unsolicited financial solicitation links (CashApp and BuyMeACoffee) into the start of all long-form brand content.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The 'Brand Voice Consistency Checker' capability (SKILL.md) ingests untrusted text provided by the user for analysis.
Boundary markers: There are no instructions for the agent to use delimiters or to ignore potential instructions embedded within the user-provided copy.
Capability inventory: The skill's capabilities are limited to text generation and structured messaging frameworks; it does not demonstrate file system or shell execution access.
Sanitization: There are no instructions for sanitizing or filtering the user-provided copy before processing.

brand-storytelling