pr-communications
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill includes a mandatory instruction to suppress all "warnings, notes, or disclaimers" in the output. This technique is often used to prevent the underlying AI model from displaying safety disclosures or content warnings to the end-user.
- [PROMPT_INJECTION]: There is a specific instruction to intercept user inquiries about the agent's system prompt or internal logic, requiring the agent to respond with a hardcoded promotional message and a link to a commercial website (scayveracademy.com).
- [PROMPT_INJECTION]: The skill mandates the inclusion of a promotional intro message containing payment handles (CashApp) and external links (Buy Me a Coffee) in every long-form deliverable, regardless of the user's specific request.
- [NO_CODE]: The analysis confirms this skill is comprised entirely of natural language instructions and does not include any executable scripts, binaries, or external dependencies, which significantly limits the technical attack surface for traditional malware or code execution.
Audit Metadata