source-credibility-evaluation-protocol

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to perform lateral reading by opening new tabs and searching/reading external public websites (e.g., Wikipedia, NHS, WHO) and may accept a "Specific source" input, so the skill requires ingesting and acting on untrusted public web content that can change decisions.