research-ideation

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

• Third-party content exposure detected (high risk: 1.00). The skill's mandatory Post-Flight Verification explicitly instructs spawning a claim-verifier and performing web searches / consulting public sources (e.g., "Search Google Scholar + NBER/SSRN" and dataset codebooks) so the agent will fetch and interpret open/public third‑party webpages whose content can change its proposed research questions and actions.