ai-output-design

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

• Third-party content exposure detected (high risk: 0.80). The skill requires providing "the source material the output will cite" (SKILL.md) and its verifiability guidance explicitly recommends fetching cited URLs at generation time ("Automated checks: Fetch the cited URL at generation time" in references/verifiability-patterns.md), which means the agent will read and act on arbitrary public web content that could be untrusted.