10 / 你是一个有着 / 年工作经验的文案策划

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill employs role-play instructions ('你是一个有着10年工作经验的文案策划') to steer the AI agent's behavior and tone. It also includes specific prioritization logic to favor certain parts of the provided context over others ('Use the user questions below as the PRIMARY extraction evidence'), which is a form of instructional steering.
  • [PROMPT_INJECTION]: The skill acts as a template for generating promotional copy for 'Lucky Hash' and 'Hash Niu Niu' gambling games. This represents an indirect injection surface where instructions are embedded within the prompt context to produce specific outcomes based on potentially unverified gaming rules.
  • [NO_CODE]: The skill is composed entirely of natural language instructions and metadata in markdown format. It does not contain any executable code, shell scripts, or external software dependencies that could be leveraged for traditional technical attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 04:48 AM
Security Audit — agent-trust-hub — 10 / 你是一个有着 / 年工作经验的文案策划