fact-checker

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md process explicitly requires researching claims "using authoritative sources (web_search + web_fetch)" (Process step 2), which fetches and ingests public web content that the agent will read and use to drive fact-checking decisions, exposing it to untrusted third-party content that could carry indirect prompt injection.