think-expected-value-decision-tree

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

• Third-party content exposure detected (low risk: 0.10). The required workflow takes the user's own decision, options, and probabilities, but step 3 (SKILL.md:38) directs the agent to source base rates via the companion reference-class-forecasting skill, which may involve searching or fetching arbitrary outsider-authored web content, so exposure is plausible but not mandated.