write-a-prompt

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs at runtime to "Read the current page when web access is available" for OpenAI's prompting guide (https://learn.chatgpt.com/docs/prompting), meaning external content fetched from that URL will be used as the source of truth and can directly influence generated prompts.