discovery-interview

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md Phase 3 "Research Loops" explicitly instructs the agent to "Spawn an oracle agent or use WebSearch/WebFetch" and gives examples of gathering findings from public sites (e.g., AllRecipes, Tasty, Paprika), so the agent will fetch and interpret untrusted public web content as part of its workflow which can influence subsequent decisions.