fact-checking

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow explicitly directs the agent to "identify authoritative sources" and "check each claim against at least two independent sources" (e.g., public sites like Snopes, PolitiFact, government records, peer‑reviewed papers and other web links shown in examples), which requires fetching and interpreting open/public third‑party content that can materially change the agent's verdicts and actions.